Continuing the December 2013 [1] Tip of The Month (TOTM), this tip investigates the effect of CO₂ in the feed gas, stripping gas rate and the effect of the triethylene glycol (TEG) mass circulation ratio on the TEG vaporization loss from the regenerator and contactor columns top. By performing rigorous computer simulations of a TEG dehydration process, two charts for quick estimation of TEG vaporization losses from regenerator top and contactor top, which can be used for facilities type calculations are developed. In addition, the effect of CO ₂ in the feed gas on the TEG vaporization losses for a case study is shown.

COMPUTER SIMULATION RESULTS

In order to study the effect of CO₂, stripping gas rate and TEG mass circulation ratio on the TEG vaporization losses, the TEG dehydration process was simulated using ProMax [2] software with its Soave-Redlich-Kwong (SRK) [3] equation of state (EOS). The process flow diagram used for these simulations is the same as in the November 2013 TOTM [4] and is shown in  Figure 1.

The water-saturated gas with a water content of 915 kg/10⁶ std m³ (57 lbm/MMSCF) enters the bottom of the contactor column at 37.8°C (100°F) and 6897 kPaa (1000 psia) at a rate of 2.835×10⁶ std m³/d (100 MMSCFD). The feed gas studied was either sweet or contained 10 mole % CO₂. The contactor column has three theoretical trays. The lean TEG solution enters at the top of the contactor column and flows down in the column. As shown in Figure 1, the water content of the dried gas is 10 mg/std m³ (0.63lbm/MMSCF). The rich TEG solution contains 96.1 mass percent TEG entering the still column at 100°C (212°F) and 515 kPaa (74.7 psia). The reboiler temperature was set at 204.4°C (400°F) and boil-up ratio of 0.1 (molar bases). Two theoretical trays in the regenerator (still) column (N_R = 2) and two theoretical trays (N_S = 2) in the striping gas section were utilized. The striping gas enters the bottom of the stripping gas section at 204°C (399°F) and 524 kPaa (76 psia). Methane was used for the stripping gas at a rate of 53.6 std m³/h (1893 scf/hr). The regenerated lean solution contains 99.86 mass percent TEG and the ratio of stripping gas to lean TEG liquid volume rates is 20 std m³ of gas/std m³ of lean TEG solution (2.67scf/sgal) or a mass ratio of 28.3. The regenerator (still) top temperature is 91.4°C  (196.5°F). If the same stripping gas was sparged directly into the reboiler (N_S = 0, no stripping column), with everything else remaining the same, the regenerated solution contains 99.2 mass percent TEG and the regenerator column top temperature remains practically the same and is 91.1°C  (196°F). For the above case (N_S = 2) the number of theoretical trays in the still column is increased from 2 to 3 (N_R = 3); the lean TEG concentration increased slightly from 99.6 to 99.8 mass percent but the regenerator column top temperature remained the same.

Using a similar set up as is shown in Figure 1, several simulations were performed for a range of stripping gas rates, for N_R=2, N _S=2 and reboiler pressure of 110.3 kPaa (16 psia) and temperature of 204.4°C (400°F). The results of these simulation runs are presented in Figures 2 and 3.

Figure 1. Sample results using ProMax [2] for TEG dehydration with reboiler P=110.3 kPaa (16 psia) with N_R=2 and N_S=2 [4]

Figure 2 presents the variation of the TEG vaporization losses from still/regeneartor column top with mass circulation ratio and the stripping gas rate. In addtion to the sweet feed gas (solid lines), Figure 2 also presents results for a feed gas containing 10 mole % CO₂ identified by the dashed lines.

Figure 2 indicates that as the stripping gas ratio increases the TEG vaporization losses decrease due to the decrease in temperature at the top of the regenerator column. This figure also indicates that as the TEG mass circulation ratio increases, the TEG vaporization losses increases initially due to lower water content of TEG solution, followed by a decreasing trend with the exception for the case of low stripping gas rate. Figure 2 also indicates that the presence of 10 mole % CO ₂ in the feed gas has little effect on the TEG vaporization losses from the regenerator column top. Figure 2 can be used for a quick estimate of the TEG vaporization loss from regenerator top for a given stripping gas rate and TEG circulation mass ratio.

Figure 2. Variation of TEG vaporization loss from regenerator top with circulation mass ratio and stripping gas rate at top P=101.3 kPaa (14.7 psia) and reboiler P=110.3 kPaa (16 psia) at 204.4°C (400°F)

As expected Figure 3 indicates that the TEG vaporization loss from the contactor top increases slightly with the stripping gas rate. In addition, this figure shows that as the TEG mass circulation ratio increases beyond 15 mass of TEG/mass of water removed, the TEG losses remain almost constant. The lower mass circulation ratio corresponds to the lower amount of TEG in contractor and lowers vaporization losses.

Figure 3 also shows that the TEG vaporization loss is higher for the feed gas containing 10 mole % CO₂(dashed lines) by a factor of about 18%.

The comparison of Figures 2 and 3 indicates that the TEG vaporization loss from the contactor top is almost 10 times higher than the loss from still/regenerator column top.

Figure 3. Variation of TEG vaporization loss from contactor top with circulation mass ratio and stripping gas rate

CONCLUSIONS

This TOTM studied the effect of CO₂ in the feed gas, mass circulation ratio, and stripping gas rate on the TEG vaporization losses from the contactor top and regenerator top. The charts exhibited a quick estimation of the TEG vaporization losses from still/regenerator column top and contactor top at a specified stripping gas rate and TEG mass circulation ratio to achieve the desired level of lean TEG concentration. These are based on the rigorous calculations performed by computer simulations and can be used for facilities type calculations for evaluation and troubleshooting of an operating TEG dehydration unit. In addition, the following observations were made for the cases studied in this TOTM:

>>The TEG vaporization loss from the contactor top is almost 10 times higher than still/regenerator column top (see Figures 2 and 3).

>>Presence of 10 mole % CO2 in the feed gas to contactor column increases the TEG vaporization loss from the top of contactor columns (Figures 3) but has a small effect on TEG vaporization loss from the regenerator column (Figure 2).

>>The results of vaporization loss from the flash tank separator were very small, in the order of 0.0025 lit of TEG/10⁶ Std m³ of gas (0.00002 gal of TEG/MMscf of gas).

>>Though not studied in this TOTM, mechanical losses such as entrainment from contactor top and regenerator top, as well as leaks from pump seals are usually much higher than the vaporization losses presented here.

To learn more about similar cases and how to minimize operational problems, we suggest attending our G4 (Gas Conditioning and Processing), G5 (Practical Computer Simulation Applications in Gas Processing), and G6 (Gas Treating and Sulfur Recovery) courses.

PetroSkills offers consulting expertise on this subject and many others. For more information about these services, visit our website at http://petroskills.com/consulting, or email us at consulting@PetroSkills.com.

By: Dr. Mahmood Moshfeghian

References

Moshfeghian, M., http://www.jmcampbell.com/tip-of-the-month/2013/12/estimating-teg-vaporization-losses-in-teg-dehydration-unit/, Tip of the Month, December 2013.

ProMax 4.0, Bryan Research and Engineering, Inc., Bryan, Texas, 2017.

Soave, G., Chem. Eng. Sci. Vol. 27, No. 6, p. 1197, 1972.

Moshfeghian, M., http://www.jmcampbell.com/tip-of-the-month/2013/11/estimating-still-column-top-temperature-in-teg-dehydration-unit/, Tip of the Month, November 2013.

Moshfeghian, M., http://www.jmcampbell.com/tip-of-the-month/2013/09/high-pressure-regeneration-of-teg-with-stripping-gas/, Tip of the Month, September 2013.

In the February and May 2010 tips of the month (TOTM) we presented the distribution and concentration of sulfur compounds in an NGL Fractionation Train (NFT) using commercial simulation software [1-3].

Sulfur compounds that may be present in NGL include H2S, COS, CS2, and mercaptans (RSH). These are typically present in low concentrations and are often removed by adsorption on a molecular sieve, reaction with a caustic solution, or amine treating. In addition, the methanol content of the liquid hydrocarbon can be as high as 0.6 mole % (GPA-Midstream RR 149 [4]).

In this TOTM we will present the effect of methanol on the distribution and concentration of the sulfur compounds in the same NFT using ProMax [5] based on the Peng-Robinson equation of state (PR EoS) [6]. The software’s built-in binary interaction parameters were used. The tabular and graphical simulation results are presented. The feed composition, rate, and condition are shown in Table 1.

Table 1. Feed composition and condition [3]

The NFT process flow diagram is shown in Figure 1 [3]. The column and product specifications are shown in Table 2. An overall tray efficiency of 90 % was used for all columns.

Figure 1. Process Flow diagram for NGL Fractionation Train (NFT) [3]

Table 2. Column specification used in simulation [3]

Expected Product Distribution:

Figure 2, like Figure 9 presented by Likins and Hix [7], shows a descending order log scale bar-graph of the pure compounds vapor pressure for the components of interest to this study. The vapor pressures shown in this diagram are calculated using ProMax [5].

This figure shows that COS should distribute to both the ethane and the propane streams. MeSH, with a vapor pressure close to n-butane should distribute primarily with the butanes with a small amount distributing to the pentane stream. EtSH, having a vapor pressure between butane and pentane, should distribute primarily with butane and pentane. CS2 should distribute primarily to the pentane and the C6+ streams with only minor distribution to the butane stream. The heavier sulfur compounds should end up almost entirely in the C6+ stream.

Figure 2. Pure compound vapor pressure at 12.8°C (55°F)

Results of Computer Simulation:

The NFT described in the previous section was simulated using ProMax [5] based on the Peng-Robinson equation of state (PR EoS) [6]. In this study, the built-in (library) binary interaction parameters were used even though we recommend evaluating the accuracy of VLE results against experimental data and if necessary the insertion of VLE data regression into the EoS interaction parameters. This regression may be required to adequately model the systems dealing with mercaptans.

The focus of this study is on the distribution (% recovery) and concentration (PPM) of the sulfur compounds in the product streams in the absence and presence of methanol. Table 3 presents the PPM concentration of sulfur compounds in the feed and product streams. The maximum concentration of each compound in the product streams are shown in red color fonts.

Table 3. Concentration (ppm, mole) of sulfur compounds in the gas and product streams

Figures 2 through 6 present bar-graphs of the recovery of each sulfur compound in the “Gas” and the other product streams. Note in these figures, the “Gas” represents the sum of gas streams 9, 13, and 16. The mole percent recovery is defined as the number of moles of a component in the product stream divided by the moles of the same component in the feed stream (Stream 5).  Figures 8 through 11 present the effect of methanol on the recoveries of the sulfur compounds in the product streams.

H2S and COS:

Figure 3 shows the distribution and recovery of H2S and COS in the Gas, C2 and C3 product streams. As expected, much of the H2S distributes in the gas and the C2 product streams and much of the COS ends up in the C3 product stream. Note H2S in C3 ~ 0.

MeSH: Figure 4 shows the distribution and recovery of MeSH in the Gas, C2, C3, C4, and C5 product streams. Most of the MeSH distributes to the C3 and C4 stream. Note MeSH in C2 and C5 is ~ 0.

Figure 3. Distribution and recovery of H2S and COS in the gas, C2 and C3 product streams (H2S in C3 ~ 0)

EtSH and CS2: Figure 5 shows the distribution and recovery of EtSH and CS2 in the C3, C4, C5, and C6+ product streams. Most of the EtSH ends up in the C4 and C5 streams and most of the CS2 ends up in the C5 stream which are consistent with Figure 1 data. Note EtSH in C3 and C6+ is ~ 0 and CS2 in C6+ is ~ 0.

iC3SH and iC4SH: Figure 6 shows the distribution and recovery of iC3SH and iC4SH in the C4, C5 and C6+ product streams. As expected, iC3SH ends up in the C5 and C6+ streams and all of the iC4SH ends up in C6+ stream. Note iC3SH in the C4 is ~ 0.

Figure 4. Distribution and recovery of MeSH in the Gas, C2, C3, C4, and C5 product streams (MeSH in C2 and C5 is ~ 0)

Figure 5. Distribution and recovery of EtSH and CS2 in the Gas, C2, C3, C4, C5, and C6+ product streams (EtSH in C3 and C6+ is ~ 0 and CS2 in C6+ is ~ 0)

Figure 6. Distribution and recovery of iC3SH and iC4SH in the C4, C5, and C6+ product streams (iC3SH in the C4 is ~ 0)

Effect of Methanol:

To investigate the effect of methanol on the recoveries of sulfur compounds in product streams, the zero concentration of methanol in the feed (stream 5) was changed to 0.01, 0.1, 0.5, and 1 mole %.

As shown in Figure 7, simulation results indicated that almost all of the methanol was distributed only between C4 and C5 product streams. Based on Figure 2, all the of methanol should with go with C6 and C7 but it is not from the same family. Figure 7 also indicates that the level of recoveries in these two product streams is a linear function of the methanol concentration in the feed stream.

Figure 7. Recovery of methanol in product streams vs. its concentration in the feed

The simulation results indicated that the recoveries of H2S, COS, and iC4SH are independent of methanol concentration in the feed. The effect of methanol concentration in the feed on the recoveries of the other sulfur compounds are presented in Figures 8 through 11.

Figure 8 indicates that the presence of methanol in the feed has negligible effect on the recoveries of MeSH in the product streams.

Figures 9 and 10 indicate that the presence of methanol in the feed has appreciable effect on the recoveries of EtSH and CS2 in the C4 and C5 product streams but negligible effect on the C6+ product stream, respectively.

Figure 11 indicates that the presence of methanol in the feed has considerable effect on the recoveries of iC3SH in the C5 and C6+ product streams but negligible effect on the C4 product stream.

Figure 8. MeSH recovery in the Gas, C3, and C4 product streams.

Figure 9. EtSH recovery in the C4, and C5, and C6+ product streams

Figure 10. CS2 recovery in the C4, and C5, and C6+ product streams

Figure 11. iC3SH recovery in the C4, and C5, and C6+ product streams

Conclusions:

The calculation results presented and discussed here are specific to the NGL fractionation train considered in this work, but there are some general conclusions that can be drawn from this study. The results can be summarized as follows:

• The highest concentration of MeSH occurs in the C4 product (stream 20).
• The highest concentration of EtSH occurs in the C5 product (stream 23).
• The highest concentration of CS2 occurs in C5 product (stream 23).
• The highest concentration of iC3SH occurs in C5 Product (stream 23).
• The highest concentration of iC4SH occurs in C6+ Product (stream 24).
• Any methanol in the feed stream will be distributed between C4 and C5 product streams.
• The amount of methanol recoveries in the C4 and C5 product streams is a linear function of the methanol concentration in the feed stream.
• Presence of methanol in the feed has no effect on the recoveries of H2S, COS (the most volatile) and iC4SH (the least volatile) in the product streams.
• Presence of methanol in the feed has little effect on the recoveries of MeSH in the product streams.
• Presence of methanol in the feed has considerable effect on the recoveries of EtSH, CS2 and iC3SH in the product streams. In the absence of methanol, most of these three compounds were distributed in C4 and C5 product streams; therefore, their concentration in these two product streams are affected by presence of methanol.

The binary interaction parameters used in the EoS play an important role in the VLE behavior of the system under study, and affect the distribution of the sulfur compounds present in the feed. Use of improper or incorrect binary interaction parameters may generate erroneous results. Care must be taken to use correct values of binary interaction parameters. In this work, the simulator library values of the binary interaction parameters were used.

The results also indicate that some of these compounds were not distributed among the hydrocarbon products in the same way one would expect from their volatilities and concentrations. This may be explained by the conclusion reported by Harryman and Smith [8, 9] who wrote “iC3SH is formed during fractionation within the depropanizer and the deethanizer.” It should be also noted that the VLE behavior of components in the mixture is not the same as each individual component behavior. Therefore, further evaluation should be conducted to arrive at a concrete decision. This should be a good reason to perform laboratory tests and detailed thermodynamic calculations to determine process flow rates and composition. Detailed process analysis should always be made to justify and prove correct decisions as to selection of process flow schemes.

To learn more about similar cases and how to minimize operational problems, we suggest attending our G4 (Gas Conditioning and Processing), G5 (Practical Computer Simulation Applications in Gas Processing), and G6 (Gas Treating and Sulfur Recovery) courses.

PetroSkills offers consulting expertise on this subject and many others. For more information about these services, visit our website at http://petroskills.com/consulting, or email us at consulting@PetroSkills.com.

By: Dr. Mahmood Moshfeghian

Interested in receiving Tip of the Month emails? Sign up today!

Reference:

1. Moshfeghian, M. “Distribution of Sulfur-Containing Compounds in NGL Products”, TOTM, Feb 2010.
2. Moshfeghian, M. “Distribution of Sulfur-Containing Compounds in NGL Products by Three Simulators,” TOTM, May 2010.
3. Al-Sayegh, A.R., Moshfeghian, M.  Abbszadeh, M.R., Johannes, A. H. and R. N. Maddox, “Computer simulation accurately determines volatile sulfur compounds,” Oil and Gas J., Oct 21, 2002.
4. Gas Processors Association, “GPA RR-149: Vapor-Liquid and Vapor-Liquid-Liquid Equilibrium for H2S, CO2, Selected Light Hydrocarbons and a Gas Condensate in Aqueous Methanol or Ethylene Glycol Solutions,” 1995.
5. ProMax 4.0, Bryan Research and Engineering, Inc, Bryan, Texas, 2017.
6. Peng, D.Y. and D. B. Robinson, Ind. Eng. Chem. Fundam. 15, 59-64, 1976.
7. Likins, W. and M. Hix, “Sulfur Distribution Prediction with Commercial Simulators,” the 46th Annual Laurance Reid Gas Conditioning Conference Norman, OK 3 – 6 March, 1996.
8. Harryman, J.M. and B. Smith, “Sulfur Compounds Distribution in NGL’s; Plant Test Data – GPA Section A Committee, Plant design,” Proceedings 73rd GPA Annual Convention, New Orleans, Louisiana, March, 1994.
9. Harryman, J.M. and B. Smith, “Update on Sulfur Compounds Distribution in NGL’s; Plant Test Data – GPA Section A Committee, Plant design,” Proceedings 75th GPA Annual Convention, Denver, Colorado, March, 1996.

This TOTM will discuss the phenomenon of infinite recycles and their impacts on design, troubleshooting and optimization to increase oil production simply by adjusting setpoint. As well as flaring reduction, unconventionals, identification and solutions.

Question: If you have an infinite recycle of LPG in your gas processing plant, how much horsepower does your compression system need?

Answer: Infinite

Liquefied Petroleum Gas (LPG) is composed of propane and butane. The pressure at which LPG becomes liquid, called its vapor pressure, varies depending on composition and temperature; for example, it is approximately 220 kPa (32 psi) for pure butane at 20 °C (68 °F), and approximately 2,200 kPa (320 psi) for pure propane at 55 °C (131 °F).

In Figure 1 you have the standard propane refrigeration loop. Compressor (K-100) / Cooler (E100) / Separator (V-100) / JT Valve (VLV-100)/ Chiller Load (E101) and recycle back to the compressor.  Under normal operation, stream No 4 has zero flowrate and the system is a closed loop.  If we were to add an additional stream of pure propane to the system at the compressor suction it creates an infinite loop since the vapor out of V100 will still be zero due to the thermodynamic limits of the phase diagram. This system requires infinite horsepower and the rest of the equipment will need to increase in size as well. The molecules are trapped in a loop.

Figure 1 – Propane Refrigeration Loop

Steady State Process simulators generally handle this situation with an unconverged solution or in some instances negative flowrates.  Why is this important?  Although the simulator will not allow the condition to occur.  It can occur during actual operation.

Figure 2 shows a typical three stage compression system, it is like the propane refrigeration system process line up.  In a typical design where are the compressor discharge scrubber liquids sent?  Generally, it’s the previous stage suction scrubber.  Eventually it makes its way to the dry oil tank.  If the material in the scrubbers has propane and butane as a liquid and then sent to an essentially atmospheric tank the liquids will flash, and then be processed by the Oil Stock Tank Vapor Recovery compressor which will send the molecules back to the compression system.  This can set up an infinite recycle or very large recycle of LPG in the compressor system.

Figure 2 – Typical three stage compression system – notice a similar process in the boxes.

As shown in Figure 3 the system response to this recycle of LPG from compressor scrubbers will be to load the system to its capacity, and then flare.  This flaring breaks the recycle, but at a cost.

Indications that this is occurring at your offshore platform, gas plant, or unconventional tank battery is a hot flare with a smoky tail.  LPG’s have a significantly higher heating value than methane and the smoke is being caused by high temperatures in the flame cracking the LPG, and creating soot.  The flame will be a deep red rather than orange and we feel higher radiation from flare when walking in the area.

Figure 3 – System response to an LPG infinite recycle will be to completely load the system and then flare the excess gas.

Figure 4 shows how the flash gas composition becomes richer and richer in LPG.  This can lead to infinite recycles in the last two separators of the oil stabilization system. The last stage flash has 47% LPG in the vapor flash gas.

Figure 4 – Impact of stage separation on flash gas composition [1]

When we are in the design phase in field development planning gas compositions have a degree of uncertainty.  It is wise to have multiple dispositions for these LPG streams. They need a way out: sales gas, oil or fuel.

DESIGN SOLUTIONS:

1. Pump interstage liquids into the sales gas using a positive displacement pump.  The limitation here is the maximum heating value of the sales gas.  Liquids may need to be gas stripped of water to avoid hydrates.

2. Pump interstage liquids into the stock tanks oil.  The limitation is crude vapor pressure and additional recycling.

3. Operate your interstage compressor discharge coolers at a high enough temperature to avoid liquid accumulation in the discharge scrubbers.  This option increases the horsepower requirements of the next stage since you are sending higher temperature gas forward to the next stage of compression.  This is also a method to deal with recycles and reduce flaring in operation, but will reduce the inlet oil production since you will be using additional compression HP by operating at higher temperatures.

4. Fuel gas – spike the LPG into the low-pressure fuel gas system.  This is a preferred option, but the direct fire equipment must be designed to accommodate the higher heating value.

5. Change the gas composition by adding hydrocarbon dry stripping gas to the crude oil.

6. If unable to send the molecules to the gas (due to Btu limit) or oil (RVP limit) then a separate disposition like Pipeline. Pressurized trucks or a tanker is needed.

7. Add another process – in offshore consider power generator or low salinity waterflood- generation of distilled water using multi-effect evaporation. (e.g. Mechanical Flare)

Bottom line is that all the molecules need an evacuation route. 

DESIGN CASE No 1:  Offshore Platform Associated Gas Processing

The case in Figure 5 is taken from “Oilfield Processing of Petroleum-Volume 1- Natural Gas” [2].  No recycling of LPG occurs as all liquids from compressor discharge scrubbers are routed to the sales oil pipeline.

Figure 5 – All compressor discharge scrubbers directly sent or pumped into the sales crude pipeline [2]

CASE STUDY No 2:  NGL Recovery Plant with 50,000 bbl Floating Roof Tank Emissions Issues

The NGL recovery plant (Figure 6) processed a small amount of crude oil and was spiking liquids into the crude.  An infinite recycles developed and the stock tanks were degassing large volumes of LPG.  Notice the large temperature drop of the crude going from 145°F (63°C) to 102°F (39°C) as the fluid flashes into the last stage separator. The design intent was to maximize oil volume and eliminate discharge scrubbers, by utilizing the main train separators.

This recycle was broken by adding hydrocarbon dry residue gas from the NGL plant at the inlet to the Electrostatic Treater mixing with the crude oil, and stripping out the LPG breaking the LPG recycle. (Same location near DeC4 bottoms injection point) The blue and red lines show the paths for the LPG Traps.

Figure 6 – Operational problems caused by infinite recycles / butane trap in an oil processing train leading to floating roof storage tank degassing.

CASE STUDY No 3:

An offshore FPSO (Floating Production, Storage and Onboarding) in the North Sea was in operation for two years and its production was constrained by its compression system.  The flare had a smoky tail.

Question: What would you recommend looking for?

Answer: Infinite Recycles.

After some process simulations, the pressure in the last stage of oil separation was increased by 0.5 barg (7.25 psig).  This broke the internal infinite recycle of LPG in the oil dehydration train sending the LPG into the crude oil rather than the flash gas to the compressor.  The resulting oil was within contract vapor pressure limits once cooled.

The upside was that inlet oil production was increased 30% immediately with zero capex!

Lesson- you need to visit the site to identify opportunities.  And if the system is running it doesn’t mean there is no opportunity to improve.

UNCONVENTIONAL DESIGNS:

Many unconventional plays have standard tank batteries with Tank Vapor Recovery Systems and/or Thermal Oxidizers.  Numerous plays have very rich associated gas that form these infinite recycles in their flash gas compression and tank vapor recovery systems.  Simply sending the liquids back to the stock tanks or water tanks will not fix this problem.  The molecules must be sent forward to the gas gathering system.  Also, ensure your design will function in the winter and that LPG condensation in suction piping will not occur, and that stock tanks are insulated and heated to prevent reflux condensation in the stock tanks.

QUICK TIPS:

Do not design for a single point of operation. (i.e. 40,000 bopd and 100 MMSCFD)

To ensure that the desing will work make sure to design during the early start-up phase with high or low rates. Additionally, make sure to design for extreme ambient condition variations in Summer/Winter and Day/Night. The design should also work during shutdown and composition changes over the life of the development.

SUMMARY:

Recycles limiting production are frequently found in oil processing facilities worldwide. However, they are easy to identify and fix with setpoint adjustments to increase production. Now that you know how to identify, prevent and take advantage of it for your company…go out and make some process improvements to your systems and make your company more profitable!

To learn more about similar cases and how to minimize operational problems, we suggest attending our G4 (Gas Conditioning and Processing), G5 (Practical Computer Simulation Applications in Gas Processing), PF3(Concept Selection and Specification of Production Facilities in Field Development Projects), PF4 (Oil Production and Processing Facilities), and PF49 (Troubleshooting Oil & Gas Processing Facilities) courses.

PetroSkills offers consulting expertise on this subject and many others. For more information about these services, visit our website at http://petroskills.com/consulting or email us at consulting@PetroSkills.com.

By: James F. Langer, P.E.

BETX stands for benzene, ethylbenzene, toluene, and xylene, a group of compounds that all belong to the broader category of Hazardous Air Pollutants (HAPs). Benzene and ethylbenzene are known carcinogens, and have also been shown to cause blood disorders, impact the central nervous system and the reproductive system.  Additionally, Toluene may affect the reproductive, respiratory and central nervous systems.  Xylene may have respiratory and neurological effects as well [1]. BETX components are present in many natural gas streams and are absorbed by the solvent in glycol dehydration and amine sweetening units.

In gas dehydration service, triethylene glycol (TEG) will absorb limited quantities of BETX from the gas. Based on literature data, predicted absorption levels for BETX components vary from 5-10% for benzene to 20-30% for ethylbenzene and xylene [2]. Absorption is favored at lower temperatures, increasing TEG concentration and circulation rate. The bulk of absorbed BETX is separated from the glycol in the regeneration unit and leaves the system in the regenerator overhead stream.

The emission of BETX components from glycol dehydration units is strictly regulated in most countries.  In the U.S., benzene emissions are limited to 1 ton/year (900 kg/year).  Mitigation of BETX emissions is an important component in the design of a dehydration system. Correctly estimating the quantity of absorbed BETX and understanding the factors that affect absorption levels is critical.

Moshfeghian and Hubbard [3] simulated a contactor column with two theoretical stages and a natural gas feed containing methane through n-decane (C₁ through C₁₀) and BETX compounds. The BETX concentrations in the feed gas were 400, 100, 50, and 50 ppmv for benzene, toluene, ethyl benzene, and o-xylene; respectively. The concentration of the lean TEG stream was 99.0 weight % TEG, and it was assumed the lean TEG temperature was the same as the feed gas. The feed gas was saturated with water at feed conditions. For each contactor pressure and temperature, the lean TEG circulation ratio was varied. To perform simulation, they used ProMax [4] with the Soave-Redlich-Kwong [5] equation of state (SRK EOS). Based on their simulation results, they presented two diagrams for quick estimation of BETX absorption in TEG dehydration process. Figures 1 and 2 [3] present the BETX absorption % (on weight basis) as a function of TEG circulation ratio and temperature for two pressures of 300 and 1000 psia (2069 and 6897 kPa).

This tip demonstrates application of these diagrams for quick estimation of BETX absorption in a case study. Specifically, TEG dehydration of a natural gas stream containing BETX compounds was considered and the estimated annual mass rate of BETX compounds absorbed by TEG solution. The graphical and hand calculation results were compared with the ProMax [4] simulation results and good agreement was observed.

Finally, an overview of the most commonly used designs for mitigation of BETX emissions will be provided.

.

Figure 1. Approximate BETX Absorption in TEG vs circulation ratio and contactor temperature at 300 psia (2069 kPa) [3]

Figure 2. Approximate BETX Absorption in TEG vs circulation ratio and contactor temperature at 1000 psia (6896 kPa) [3]

Example Problem

100 MMscfd (2.832×10⁶ Sm³/day) of natural gas with a composition shown in Table 1 and saturated with water at feed conditions enters a TEG contactor. The inlet temperature is 110°F (43.3°C) and pressure is 1000 psia (6897 kPa). The outlet water dew point spec is 20°F (-6.7°C). The reboiler is operating at 360°F and there is one theoretical stage of random packing in the stripper.

Determine the quantity of BETX in lb_m/year (kg/year) which would be vented to the atmosphere if no remedial actions were taken.

Assumptions: Approach temperature = 18°F (10°C) and lean TEG temperature = feed gas temperature

Solution

Based on the procedure outlined in Chapter 17 of Volume 2 of Gas Conditioning and Processing Book [6], the following parameters were determined:

Using Figure 3 for the lean TEG circulation ratio of 0.17 gpm TEG/MMscfd of gas (1.36 m³/h TEG/10⁶ Sm³/day of gas), BETX absorption weight % for temperatures of 95 and 122°F (35 and 50°C) are presented in Table 2. Note that for 122°F (50°C), the lines were extrapolated by dotted lines for lower flow rates.  By linear interpolation, the estimated values of absorption weight % for temperature of 110°F (43.3°C) are presented in the last column of Table 2.

Figure 3. Estimating BETX Absorption (weight %) in TEG for example at psia (6896 kPa) [3]

For lean TEG concentration of 99.5 weight % and circulation rate of 17 gpm (3.861 m³/hr), the BETX absorption weight % from ProMax simulation results are shown in Table 3. As shown in Tables 2 and 3, there is a good agreement between the estimated BETX absorption weight % from Figure 3 and the ProMax simulation results. The ProMax calculated dry gas water dewpoint was also 20°F (-6.7°C) that matches the specified value.

The annual mass rate of each BETX compound entering the contactor in lb_m/yr (kg/yr) can be determined from the following equations.

In the above equations q = 100 MMscfd (2.832×10⁶ Sm³/day), MW (molecular weight) and ppmv (concentration) are given in Table 4. For each BETX compound, the absorbed mass rate of is determined by multiplying mass rate entering the contactor times the corresponding absorption weight % divided by 100.  The calculation results are presented in Table 4. The last two columns of Table 4 present the annual mass rate of each component absorbed in the TEG and subsequently regenerated and vented with the water.

The bulk of absorbed BETX will be vented with the water vapor at the top of the regenerator. The most common emission mitigation strategies are to [3]:

1. Condense the regenerator overhead vapor in a partial condenser and combust the remaining vapor.  The uncondensed vapors are typically routed to an incinerator or, if a direct-fired reboiler is used, routed to the reboiler fuel gas. The liquid hydrocarbons are collected and disposed of by blending into a crude oil or condensate stream.  The condensed water is typically routed to produced water disposal.

2. Route the regenerator overhead vapors to another process stream in the facility. This is typically a low-pressure stream such as flash vapors from the last stage of a crude or condensate stabilization system.

From an operational point of view, minimizing circulation ratio is the most effective way of decreasing the absorption of BETX components. This also minimizes reboiler duty and the size of the regeneration skid. Lower TEG circulation rates require more theoretical stages in the contactor to meet outlet water content specifications, but the additional cost of a taller contactor is often offset by savings in the regeneration package. Care should be taken that the glycol circulation rate is sufficient to ensure adequate liquid distribution over the packing. Packing vendors can provide minimum circulation guidelines.

Summary

Figures 1 and 2 present a simple tool for quick estimation of absorption of BETX compounds in TEG dehydration process. For the case study considered in this tip, the estimated absorption weight % for each of BETX compounds matched well with the ProMax simulation results.

To learn more about similar cases and how to minimize operational problems, we suggest attending our G4 (Gas Conditioning and Processing), G5 (Advanced Applications in Gas Processing),  and PF49 (Troubleshooting Oil & Gas Processing Facilities) courses.

PetroSkills offers consulting expertise on this subject and many others. For more information about these services, visit our website at http://petroskills.com/consulting, or email us at consulting@PetroSkills.com.

In the past tips of the month (October, November, December 2007, January 2011, February, September 2014. And April 2015), we studied in detail the water phase behaviors of sweet and sour natural gases and acid gas systems. We evaluated the accuracy of different methods for estimating the water content of sweet and sour natural gases as well as for acid gas systems. In addition, correlations to estimate vapor water content of lean sweet and sour gases were presented.

For normal gas conditioning processes, the water vapor content of natural gases in equilibrium with water is commonly estimated from McKetta and Wehe [1] based charts like Figure 6.1 of Campbell book [2] or Figure 20.4 of Gas Processors and Suppliers Association [3]. This tip presents a chart for estimating methane gas water content at low temperatures, e.g. cryogenic processes.

Hydrate formation is a kinetic (time dependent) process. During this transient “hydrate formation period” the liquid water present is termed “metastable liquid.” Metastable water is liquid water that, at equilibrium, will exist as a hydrate. At temperatures below the hydrate temperature of the gas, the “condensed” phase will be a solid (hydrate). The water content of a gas in equilibrium with a hydrate will be lower than equilibrium with a metastable liquid. The water content of gases in the hydrate region is a strong function of composition. Page 153 of Campbell book [2] provides more detail.

Figures 1A (SI, international system of units) and 1B (FPS field system of units) present a chart for methane gas water content for temperature range of 0 °C to -80 °C (32 °F to -110 °F) and pressures of 0.345 MPa, 3.45 MPa, and 6.9 MPa (50 psia, 500 psia, and 1000 psia). The chart is based on the experimental data reported in GPA-Midstream RR 187 [4] for pressures of 3.45 and 6.9 MPa (500 and 1000 psia) and the results generated using ProMax [5] freeze model (Solid-Vapor-Equilibrium) at low pressure of 0.345 MPa (50 psia).

Figure 1A (SI). Low temperature methane gas water content
(Dashed lines are GPA-Midstream RR 187 data [4], solid line predicted using ProMax [5])

Figure 1B (FPS). Low temperature methane gas water content
(Dashed lines are GPA-Midstream RR 187 data [4], solid line predicted using ProMax [5])

Based on the methane hydrate formation diagram shown in Figure 2, the above-mentioned low temperatures are generally below the hydrate formation temperature at the corresponding pressures. Therefore, the water contents reported in Figures 1A and 1B are in equilibrium with hydrates.

Figure 2. Methane hydrate formation temperature as a function of pressure

Summary

Figure 1 presents a simple chart to estimate methane gas water content in equilibrium with hydrates. The water content of gases in the hydrate region is a strong function of composition. Where experimental data is unavailable, utilization of an EOS-based correlation that has been tuned to empirical data can provide an estimate of water content in equilibrium with hydrates.

To learn more about similar cases and how to minimize operational problems, we suggest attending our G4 (Gas Conditioning and Processing), G5 (Practical Computer Simulation Applications in Gas Processing), and PF49 (Troubleshooting Oil & Gas Processing Facilities), courses.

PetroSkills offers consulting expertise on this subject and many others. For more information about these services, visit our website at petroskills.com/consulting, or email us at consulting@PetroSkills.com.

By: Dr. Mahmood Moshfeghian

Reference:

1. McKetta, J. J., and Wehe, A. H., “Use This Chart for Water Content of Natural Gases,” Petroleum Refiner (Hydrocarbon Processing), Vol. 37, No. 8, p. 153, August 1958.
2. Campbell, J.M., Gas Conditioning and Processing, Volume 1: The Basic Principles, 9^th Edition, 2^nd Printing, Editors Hubbard, R. and Snow–McGregor, K., Campbell Petroleum Series, Norman, Oklahoma, 2014.
3. GPSA Engineering Data Book, Section 20, Volume 2, 13^th Edition, Gas Processors and Suppliers Association, Tulsa, Oklahoma, 2012.
4. Song, K. Y., Yarrison, M., Kobayashi, R., and W. Chapman, “Low Temperature V-L-E Data for Water, CO2, and Light Hydrocarbon Systems,” Gas Processors Association Research Report RR 187 Tulsa, Oklahoma, 2005.
5. ProMax 4.0, Bryan Research and Engineering, Inc., Bryan, Texas, 2017.

This tip of the month (TOTM) discusses how to “avoid a costly design error found in most oil dehydration systems worldwide”.

• Are you having reverse emulsion problems?
• Are you having problems meeting your discharge oil in water specifications.

An area to look for immediately is the Process Flow Diagram (PFD).  Normally systems work as they were designed. In cases where I have visited sites experiencing problems…the problem is the design.

Why does this occur? The world we deal with is more complex than process simulation models. They are simply thermodynamic models to predict the phase separation of multicomponent systems and perform heat and energy balances. They are useful and very well proven tools. Most designs however; do not consider the various chemicals, presented in in Table 1, that are added to a system on a routine basis.

Table 1. Example of various problems and their typical treatment

Most of these chemicals are added to a system to improve a designs performance in oil water and water oil separation. Most of these chemicals are used in parts per million (ppm) fashion using proprietary formulations from Chemical Companies. Most assets will spend $5-$10 Million dollars per year on the use of these formulations. The ability to add the chemicals is included in most designs, but their effects are not normally included in the designs and it’s left to Field Operations and the Chemical Companies to come up with a solution.

So, What’s the design error? It’s mixing fluids that have come into contact with both oil treating and water treating chemicals and then recycling them back to the front of the process.

What are these streams?

• Oil skims from WEMCOs (induced gas floatation manufacture by Western Equipment Manufacturing Company) or Flotation Cells are routinely re-routed back into the Oil Dehydration Train.
• Hydrocyclone Oil Skims
• Closed Drain Oil Skims

I would encourage you to check this out at your asset using the real fluids. I perform this demonstration during the Petroskills PF-3 Course. I take a 50% water cut sample of a 45 API Gulf of Mexico Crude Oil and produced water. I shake it up and the emulsion resolves in less than 30 seconds back into oil and water. I then add 5 -10 ml of Demulsifier and Water Clarifier being used on the Platform. Shake this up and voila… instant chocolate milk emulsion. This emulsion will not resolve in days. How is the emulsion resolved? We send it to a heater treater. In class I use a coffee cup of hot water. Let it heat soak for a couple of minutes and the sample resolves into three phases as shown in Figure 1: Oil / a scrambled eggs rag layer (complex emulsion- reverse and normal) / Water. The rag layer is normally about 25% of the total sample. This rag layer is very stable, and I have had samples stay stable for over one year.

Figure 1. Oil water complex emulsion

Never send these skims back into your main process. Consider sending them directly to the sales oil LACT. Most contracts allow you 3-5% BS&W. Use the contract to your advantage. You will not be able to break these emulsions with your existing process and they will continue to build. Some operators send them to the beach as hazardous waste for further processing. Other operators have a separate slop oil system for treating with higher temperatures, and then send streams to the sales LACT (Lease Automatic Custody Transfer), but never back into the process.

Every process needs a trash can for off spec product. Offshore it’s the closed drain tank / open drain tank. Never recycle this material back into the main oil dehydration system. Why? Let’s look at what fluids enter the closed and open drain tanks?

• Engine Lube Oils – these contain polymers that will generate emulsions
• Oxygenated water from rain. Oxygen will corrode your main production system.
• Oily Deck Drains from skid pans

How to become an Expert?

Use sources such as SPE’s One Petro. Chances are that sometime, somewhere, someone has faced the same issue you are trying to solve.  In your company is it easy to publish a paper or are there controls and bureaucracy? In most instances, it’s not easy. What does it say to you about the person who has taken the time and written a paper. They want to communicate!  What do you think their reaction will be if you contact them about their paper to get further information? They usually will be very happy to discuss the paper and give you even more insights / tips / advice / recommendations for your situation. It’s a way of leveraging your time, and getting to solutions quickly.

I once made a trip to an FPSO Offshore Brazil that was having water treating problems. They made the same design mistakes discussed in this TOTM. I searched One Petro and found another FPSO Offshore Brazil that faced similar issues and how they were solved [1]. The paper is listed in the references of this TOTM. Here is the recommendation from SPE 90409 paper:

“Oil Reject Skid. To provide a positive path for removing oily solids and chemically stabilized emulsions from the process, an oil reject skid was installed on 135 D to receive hydrocyclone reject liquids and skimming from the surge drum, as well as skimming from the sparger IGF vessels. In the oil reject vessel (Fig. 9), larger solids are removed from the cone bottom, while oily solids and chemically stabilized emulsion separates into an oil-continuous phase that is pumped off the platform without recycling back into the process. The volume of this emulsion is low, a few barrels per day, so the impact on the BS&W of crude leaving the platform is low, increasing it by approximately 0.05 %. The clarified water is returned to one of the degasser vessels, from which it reenters the water treatment process. The oil reject skid installation has been one of the major contributors to the ability of 135 D to reliably reduce the TOG of the produced water to < 20 mg/L.”

To learn more about similar cases and how to minimize operational problems, we suggest attending our G4 (Gas Conditioning and Processing), G5 (Practical Computer Simulation Applications in Gas Processing), PF3 (Concept Selection and Specification of Production Facilities in Field Development Projects), and PF49(Troubleshooting Oil & Gas Processing Facilities), courses.

PetroSkills offers consulting expertise on this subject and many others. For more information about these services, visit our website at petroskills.com/consulting, or email us at consulting@PetroSkills.com.

Reference:

1. Chifon Yand, Michel Gabrun, and Ted Frankiewicz, “Identification and Resolution of Water Treatment Performance Issues on the 135 D Platform,” SPE 90409, SPE Annual Technical Conference and Exhibition, Houston, Texas, 26-29 September, 2004.

Finding Opportunities Checklist

This Tip of the Month (TOTM) discusses places to look for opportunities to increase production quickly. I’ll discuss the concepts, techniques, and later where to look for and find opportunities….to be able to answer the Key Question: “Are we getting the most from our Asset?”

• Do you know how your asset makes money?  Is it simply oil price, and gas price?
• Do the prices of oil and gas change daily?
• Do you adjust your operating conditions based on the prices of oil and gas?
• Does fuel gas have any value?

BENCHMARKING AGAINST PHYSICS™

What are your process Key Performance Indicators?

• Energy Consumed per boe (barrel of oil equivalent) produced?
• Theoretical Energy Required per boe produced?
• Design Energy required per boe produced?

WHERE ARE THE GAPS?

How do you evaluate the performance of your:

• Pumps
• Compressors
• Processes
  • Oil / Water Dehydration
  • Gas Dehydration
  • Gas Treating

Figure 1: Technical Limit Diagram or Choke Model

Looking at the Technical Limit Diagram (Choke Model) shown in Figure 1, where could there be opportunities?

• How can we use the current equipment to increase production?
• What’s the cost of one day of downtime?
• What’s the monetary value of the annual flare volume?
• How can we increase inflow from the wells to increase production?
  • Can we reduce the inlet pressures to the facilities to increase reservoir inflow?
  • Are we holding backpressure on the facilities, or are we floating the sales gas pipeline?
  • Are there any excessive pressure drops in the flowlines/equipment?
  • Scale
  • Remove choke internals

COMPRESSORS

• Energy Efficiency:
• What’s the position of the recycle valve? Are you wasting energy?
• Reciprocating: Adjust Pockets / re-cylinder
• Centrifugal: Consider re-wheeling
• Can you reduce the suction temperature of the compressors to increase available HP
• Infinite Recycling of LPG
  • Is this occurring in your compression system?
• What’s causing downtime of your machines?
• Could your upstream separators be carrying over liquids?

HEAT EXCHANGERS

• Have the overall heat transfer coefficients become lower?
• Have the pressure drops increased?

PUMPS

•  What’s the position of your centrifugal pump discharge control valve?
  • Do you have too much HP, consider re-wheeling
• Are you shearing and creating a lot of emulsion requiring higher chemical costs?

SPECIFICATIONS

• Challenge sales specifications.
  • Are we over treating?
  • How’s our crude vapor pressure (RVP)?
• Are we over stabilizing?
  • How’s our crude BS&W?
• Are we over treating?

REDUCE OPEX

• Chemical Costs: Demulsifier
  • Are we over dosing?
  • Are we adjusting the chemical injection rate base on changes in the inlet gross rate?
  • Have we done a “shootout” with other vendors, and allowed treating to failure point?
  • Are there recycle streams we can eliminate from the process and treat separately?
  • Do we have an OPEX model?
• Have we identified our fixed and variable costs?
• What’s the cost of fuel?  Is fuel free?
• What’s the cost of flaring?
• Are there operating cost differences between “A crew” and “B crew”?
  • Set up a competition between “A crew” and “B crew”.  Bbls produced, $ energy used per bbl, $ gas flared/bbl

DOWNTIME

• How many times have we repaired this pump this year? Or over its life to date?
• What’s the cost of downtime in $/hr?  How many $ in revenue has we lost this quarter?
• How long does it take to repair?  Do we need a new seal design?  More spare parts?

CAPITAL PROJECTS

What could we add to increase production?

• Is inlet compression profitable? (Lease / Purchase)

FLARE

Is the flare smokey?

• This means its rich gas containing propane / butanes worth more than methane.  It’s a place to look for process changes to prevent the loss of these higher valued components.

WELLS

• What’s the effect of increasing the tubing size?
• What’s the effect of optimizing the gas lift?
• Do all injection wells need the same maximum pressure for gas lift or water injection? Can we segregate the system?
• Do the wells need stimulation?
• Do the wells need water shutoff?
• Manifolds – Are there leaking check valves allowing backflow from an HP well into an LP well?

Case 1: Different Molecules have different values in the Gas and Oil.  Are you maximizing the value of your products?

Offshore in the Gulf of Mexico, some contracts allow higher vapor pressure in the sales crude oil during the winter months.

KEY QUESTION: How can you take advantage of this opportunity?

Generally, selling molecules as oil has been more valuable than selling the molecules as gas.  So how can you send more of smaller LPG molecules into the sales crude oil to increase the number of crude oil barrels sold?

As you learned about vapor pressure and phase diagrams in G4 (Gas Conditioning and Processing course) you simply have to increase the pressure of the final separation, and decrease its temperature.  In winter months by lowering the oil dehydration train temperature, and increasing its pressure you increase the vapor pressure in the sales oil tank, and increase the number of barrels sold.  You can generate millions of dollars per year, by simply adjusting a temperature control setpoint and a pressure control setpoint. For Zero CAPEX.

Case 2: High Gas Prices

For a short period in 2000 a unique opportunity presented itself.  Gas Prices Spiked!

KEY QUESTION: How do we take advantage of this opportunity?

Simply send your molecules into the gas sales by dropping the oil dehydration treater pressure and raising its temperature.  Many Operators in the GOM took advantage of this opportunity. Some just sat and watched!

To learn more about similar cases and how to minimize operational problems, we suggest attending our G4 (Gas Conditioning and Processing), G5 (Advanced Applications in Gas Processing), PF3 (Concept Selection and Specification of Production Facilities in Field Development Projects), and PF49 (Troubleshooting Oil & Gas Processing Facilities), courses.

PetroSkills offers consulting expertise on this subject and many others. For more information about these services, visit our website at petroskills.com/consulting, or email us at consulting@PetroSkills.com.

By: James F Langer, P.E.

This Tip of the Month (TOTM) discusses how to increase production today, quickly, and for zero CAPEX. I’ll discuss the concepts, techniques, and where to look for and find opportunities, so we can answer the key question: “Are we getting the most from our asset?”

Many companies use benchmarking data shared anonymously with other participants in the survey. The problem is they will not have exactly the same assets. What’s being recommended here is that you benchmark you asset against physics and not others.

The good news is that no one can have better performance than the theoretical technical limit. If you determine your performance and compared it with a theoretical model / theoretical maximum and find a gap, then there’s an opportunity to improve.

BACKGROUND

You have been asked by the Asset Manager to find opportunities to increase production. In this case the well is flowing at 1200 psig (8.3 MPag) downstream of a wide open choke into a 1 mile (1.6 km) 3inch (76 mm) flowline discharging into a sales line operating at 1000 psig (6.9 MPag), and flowing gas at 10 MMSCFD (0.28×10⁶ std m³/day).

KEY QUESTION – CASE 1

Is Today a good day? Are we getting the most from this asset?

Start by comparing the following:

• What is the theoretical maximum given the current conditions – MODEL?
• What is the current operational performance – ACTUAL PERFORMANCE?

Figure 1 presents a bar chart for BENCHMARKING AGAINST PHYSICS.

The most common answer is that the well is online. We are selling gas, and the choke is wide open. There is no further investigation or action performed. This is because the gap cannot be seen without performing the engineering.

The answer from reading this TOTM should be:

• We have a process model that indicates that Benchmarking against Physics^TM the flow line should be able to deliver 13 MMSCD (0.37×10⁶ std m³/day).
• You visit the field and verify the data. The gap is a 30% increase in production, see Figure 1.
• What’s causing this? Another block valve in the system? System Scale?
• How could you increase production quickly? Remove the entire choke body internals? Replace a small ported ball valve with a full ported valve? Recommend cleaning the line? Or simply opening a valve that is partially closed at the sales connection?

What’s the incremental value of 3 MMSCFD (0.085×10⁶ std m³/day) with $4/ MMBTU ($3.8/GJ) sales gas per year? $4.35 Million USD/year. Was that worth a field trip? Absolutely!

Figure 2. Flowline Manifold – Are these valves lined up properly? Could they be leaking?

In Figure 2, what are the implications of leaking manifold valves? Could HP (high pressure) wells be sending gas backwards into lower pressure wells? Are stronger wells backing out lower pressure wells to sales?  Are a competitor’s higher pressure wells backing out our production?

KEY QUESTION – CASE 2

Is Anything wrong with our sale gas pipeline?

BACKGROUND

The pink points in Figure 3 are the gas sales in MMSCFD on the right axis, and the dark blue points are the pipeline differential pressure. The rate is decreasing over time and the pressure drop is decreasing over time.

Figure 3. SCADA DATA for pipeline performance gas rate in MMSCFD and differential pressure over time

The pipeline is a 14 inch (356 mm) 60 mile (96 km) from a major offshore asset. If the gas pipeline becomes plugged, then the oil production and gas production needs to be shut down; approximately 100,000 bopd (15,900 m³/day) and 200 MMSCFD (5.65×10⁶ std m³/day) of gas. The operational data are presented in Table 1.

Table 1- Operational data performance

A UNISIM [1] Model shown in Figure 4 was completed to determine if the current performance matched the actual operational performance.

Figure 4 – UNISIM PIPELINE PERFORMANCE

The top pipeline simulation in Figure 4 matched the operating data performance at 200 MMSCFD (5.65×10⁶ std m³/day), however; the pipeline at the lower rate of 169 MMSCFD (4.77×10⁶ std m³/day) did not match the observed performance.  In order to match the performance a choke 4 inches (102 mm) in diameter by 100 ft (31.1 m) was added to the 60 mile (96 km) 14 inch (356 mm) pipeline model.

The platform had been experiencing asphaltenes in its separators. They were lucky to have performed the engineering analysis, and have a scheduled shutdown to clean a section of the pipeline at the base of the sales gas riser.  How are you monitoring your critical cash flow pipelines?

KEY QUESTION – CASE 3

How can we improve the performance of gas pipeline using SCADA data?

Here is another quick tip using SCADA Data and Excel.

From the GPSA Data Book [2] we know that gas flowrate (Q) is proportional to a constant times the square root of the static pressure (P₁) times the differential pressure (ΔP).

So the trick is that if you construct a plot of C’ versus time… C’ is constant unless fouling has occurred. Fouling is difficult to find from just looking at data since it’s a chronic problem overtime, and not generally and acute one.

Figure 4 shows some SCADA Data using this technique. In order to scale the data, instead of C’ being plotted, the reciprocal of C’ squared was plotted to scale the plot.  This still should be a constant over time, and you can see the inflection point on the red line on this two years of data.

Figure 5. C’ Plot over time – Red line indicates a changing C’

Although this case reflected an asphaltene accumulation over time creating a localized plug. The unfavorable conditions would yield the same results as if it were a uniform scale along the entire pipeline length.

In upstream instances for subsurface models it’s difficult to see trends, because of the nature of the reservoir. We start off with high pressures, and flowrates, and these continually change over time.  The compositions oil / gas / water and their relative proportions change as well over time.  With all of these items changing overtime it’s sometimes difficult to detect small changes by simply looking at the data.  As shown in the previous examples if you Benchmark Against Physics^TM it becomes very clear what’s happening.

The last item we will add to the bar graph tool (Figure 6) is a bar for the original design capacity.

Figure 6. ACTUAL / MODEL / ORIGINAL DESIGN CONDITIONS

KEY QUESTION – CASE 4

How can we increase the Gas Plant Throughput to full capacity?

BACKGROUND

You get a call from one of your Gas Plants in Europe, and it’s having a problem. It used to operate at full capacity, but something has changed and the operators can’t get more than 50% capacity through the plant. The plant processes several Billion Standard Cubic Feet of gas per day, and at current rates will lose over a $1 Billion USD this year if it is not corrected.

You build a model to compare actual performance with theoretical, and you build the bar charts for major pieces of equipment throughout the gas plant. The plant is a straight refrigeration gas plant that has stabilization of recovered liquids. It’s been there for 20 years operating without a problem. It processes about one third of the gas coming into the country.

After looking at the model and the graphs it’s clear to you what the problem could be. I flew from Houston to London. Visited the plant. Assessed the situation, and made some setpoint changes with the operators which resolved the issue, and then flew back to Houston the next day.

Why did I take the time to visit the plant? Below is some sage advice from Norman Lieberman [3].

“The tech service engineers employed by my clients, who work with me on these jobs, always ask what special techniques I employ to solve these problems. The procedure I use is the same one I used 10 years ago:

1. Discuss the problem with the shift operators.
2. Personally collect field data and carefully observe the operation of the unit.
3. Develop a theory as to the cause of the malfunction.

The error my clients often make is that they develop a theory, usually with process computer simulations, as to the cause of the malfunction. The theory is then reviewed with management and other technical personnel at a large meeting. If no one objects to the theory, it is accepted as the solution to the problem. Typically, no one at the meeting has discussed the problem or the solution with the shift operators, nor has anyone personally observed the process deficiency in the field. Finally, the intended solution is not put to a plant test to see if it is consistent with the problem. This approach to solving refinery process problems by the major oil companies often results in wasting capital resources and engineering man-hours.”

Figure 7. Gas Plant Process Flow Schematic UNISIM

The plant schematic is shown in Figure 7, and it’s a straight refrigeration plant with a low temperature separator and liquids stabilizer. The plant must meet a gas dewpoint specification and a maximum liquids vapor pressure – RVP (Reid Vapor Pressure). Figure 8. Presents Benchmarking Against Physics bar charts for the key components in this gas plant. Note that the RVP of 6 psi (42 kPa) is well below specified value of 12 psi (83 kPa) and the plant is operating at 50% of inlet capacity.  The chiller is fully loaded.  The stabilizer overhead compressor is fully loaded. They are meeting all specs.

Figure 8

Figure 9 presents the gas plant schematic with bar charts for Benchmarking Against Physics.

Figure 9

So what’s your conclusion? The stabilizer overhead compressor is a bottleneck and current operations are over stabilizing the liquids sending too much vapor overhead (Producing a 6 psi (41 kPa) RVP versus a specification of 12 psi (83 kPa)).  If (they drop) the stabilizer bottoms temperature is reduced they will unload the stabilizer overhead compressor.  If they still are exceeding the stabilizer overhead compressor, then they can reduce the liquids recovery by not chilling the gas as much.  This will reduce the amount of liquid requiring stabilization.  The next limit is the gas dewpoint spec, and they will need to chill the inlet gas sufficiently enough to meet the specification.  Then they can begin increasing the inlet gas rate until they reach design maximum conditions.

Visiting the site, I found the plant operating as UNISIM predicted and the bar charts show. The operator’s mindset was that they needed to always chill the inlet gas to the maximum to make the company the most money. Then they backed themselves into a corner with the stabilizer operation. I went on rounds with them, and then suggest we try another approach and followed their management of change process to make setpoint changes. I explained the model results with them, and what we should expect to see as each change was slowly made. I have never shutdown a facility while making changes in 38 years because you always go slowly and are physically present to witness the cause and effects of changes. You never want to make a change and then hear the hiss of air leaving shutdown valve actuators and then have the lights go out. Going slowly builds the confidence of the operators in you and what you are doing.  Bottom line is two temperature setpoints were adjusted within limits, and the plant was returned to full capacity in 8 hours.  I then headed back to Houston for more adventures.

And as promised at the start of this TOTM…

“Increase Production Today, Quickly, and for Zero (increase in) CAPEX!”

Before you finish reading this article, please think about how to apply the concepts discussed, and TAKE ACTION in your work today. This type of work if fun and exciting…it translates directly to your company’s bottom line. Go forth, be safe, and make it a great day!

In the next TOTM I will present other places to look for optimization opportunities within your assets.

To learn more about similar cases and how to minimize operational problems, we suggest attending our G4 (Gas Conditioning and Processing), G5 (Advanced Applications in Gas Processing), PF3 (Concept Selection and Specification of Production Facilities in Field Development Projects), and PF49 (Troubleshooting Oil & Gas Processing Facilities) courses.

Reference:

1. UNISIM Design R433, Honeywell International Inc., Calgary, Alberta, Canada, 2016
2. Gas Processors Suppliers Association; “ENGINEERING DATA BOOK” 13th Edition; Tulsa, Oklahoma, USA, 2012.
3. Norman P Lieberman, N P, “Troubleshooting Process Operations,” 4^th Ed., PennWell, Tulsa, Oklahoma, 2009.

This tip is the follow up of January 2017 tip of the month (TOTM) [1] which investigated a non-refluxed condensate stabilizer column having a split design where a portion of the feed is pre-heated by heat exchange with the bottoms product (see also page 352 of Reference [2]). The remainder of the feed is fed to the top tray, similar to a standard “cold feed” stabilizer. That tip simulated the performance of a split-feed and non-refluxed condensate stabilizer column equipped with a stripping sweet gas stream to reduce the H₂S content of a sour condensate. It reported that decreasing the top feed-split from 100% to 80 %, decreases the required reboiler duty and the bottom product temperature, but increases the stabilized condensate TVP (True Vapor Pressure) and H₂S content.

This tip will perform computer simulation to study the effect of pressure on the performance of a non-refluxed stabilizer column with the same split-feed design. This tip will consider stabilizing a sour condensate for RVP (Reid Vapor Pressure) specification of 7 psi (48 kPa). The tip will report the impact of the column operating pressure on the reboiler duty, top and bottom temperatures, stabilized condensate rate, H₂S content, and TVP. The tip will present a summary of the computer simulation results and the key diagrams for the same plant.

Case Study 

Figure 1 presents a simplified process flow diagram equipped with the preheater, stripping gas and water-draw tray for stabilization and sweetening of raw sour condensates.

Figure 1. A non-refluxed stabilizer column with feed-split, side water-draw and stripping gas

The 3-phase separator upstream of the preheater removes essentially all excess/free water. The tip utilizes a feed splitter to heat up a portion of feed in the preheater by the hot stabilized condensate. The stripping sweet gas stream lowers the H₂S content of the stabilized condensate and achieves the desired condensate vapor pressure at the specified reboiler temperature. Three cases of column pressures were considered. For the comparison purposes and regardless of the specified 3-phase separator and column pressure, the separated gases from the 3-phase separator and the stabilizer column are compressed and cooled to specified pressure of 255 psia and temperature of 90 °F (1758 kPaa and 32 °C) in the compressor and cooler shown in the top-left side of this figure.

Table 1 presents the raw condensate, containing about 21 mole % H₂S, and stripping gas compositions, rates and conditions. Table 2 presents the specified column variables. The condensate stabilizer column specifications for the three cases studied are shown in Table 2. For all three cases, the 3-phase separator pressure was 5 psi (35 kPa) higher than the column feed pressure.

Table 1. Feed and stripping gas compositions, rates and conditions

Table 2. Condensate stabilizer column specifications

Simulation Results

Based on the data in Tables 1 and 2, and the process flow diagram of Figure 1, the tip performed simulation using the Soave-Redlich-Kwong (SRK) equation of state [3] in ProMax [4] software. Similar to the previous tip the bottom-splitter ratio was adjusted to meet the specified RVP of the stabilized condensate. Table 3 presents the summary of results for the three specified cases. This table indicates that decreasing the specified 3-phase separator and column pressure decreases the bottom and overhead rates, bottom and overhead temperatures, but increases the compressor power and cooler duty. At lower pressures, more gases are separated in the 3-phase separator and feed rate to the column is reduced. Therefore, the bottom product (stabilized condensate) and overhead product rates decrease. In addition, decreasing column pressure increases the relative volatility (the K-value ratio of the key components) making separation of components easier and hence the required reboiler duty decreases.

The results presented in Table 3 are practically independent of the top-feed split which varied from 80 to 100 % with an increment of 2%. The values presented in Table 3 are the calculated arithmetic averages for the top-feed split range of 80 to 100%. The simulation results for those variables which were dependent on the top-feed split are shown graphically in the following figures.

Table 3. Summary of simulation results for three specified column pressures

Figure 2 presents the required reboiler duty as a function of the top feed-split and the specified column bottom pressure. This figure indicates that decreasing column bottom pressure decreases reboiler duty and the reboiler duty increases linearly with the top feed-split %. At a specified column pressure, increasing top-feed spilt decreases the hot-feed rate to column and hence increases the reboiler duty. However, the installation of the preheater reduced the required reboiler duty by about 19% for the case of top feed-split of 80% due to the heat recovery from the hot bottom product (stabilized condensate).

Figure 3 presents the variation of stabilized condensate temperature as a function of the top feed-split. Decreasing the top feed-split from 100 to 80%, decreases the bottom product temperature by about 2 °F (1.1 °C) for the three specified bottom pressures. Decreasing the specified column pressure, decreases the bottom temperature because component vapor pressure and temperature are directly related.

Figure 4 presents the variation of stabilized condensate TVP as a function of the top feed-split and column bottom pressure. Decreasing the top feed-split, increases the stabilized condensate TVP by about 2.6 psi (18 kPa) for the three specified bottom pressures.

Figure 2. Effect of top feed-split on reboiler duty for three column bottom pressures

Figure 3. Effect of top feed-split on bottom temperature for three column bottom pressures

Figure 4. Effect of top feed-split on stabilized condensate TVP for three column bottom pressures

Figure 5 presents the variation of H₂S content of the stabilized condensate as a function of top feed-split and column overhead pressure. This figure indicates that decreasing the top feed-split from 100 to 80%, the stabilized condensate H₂S content increases by about 44, 50, and 56 ppm for bottom pressure of 195, 225, and 255 psia (1345, 1552, and 1759 kPaa), respectively. This figure also indicates that for all three cases:

• the H₂S content is below the limit of 60 ppm,
• the higher the top-feed %, the better the H₂S stripping,
• H₂S stripping is better and easier at lower pressure.

It should be noted that since the raw feed condensate leaves the feed tank (three phase separator) saturated with water but with no free water, simulation resuts showed no liquid water being trapped in the column. Therefore, no water was removed by the water-draw tray in all cases studied in this tip. Still it is advised to keep the water draw-off capabilities on the top trays.

Figure 5. Effect of top feed-split on stabilized condensate H2S content for three column overhead pressures

Conclusions

This tip investigated the impact of column and 3-phase separator pressures on the performance of a split-feed and non-refluxed stabilization column by varying the top feed-split from 80 to 100% by an increment of 2 % for three pressure specifications. Based on the simulation results, this tip presents the following observations.

Decreasing the 3-phase separator and column pressures,

1. decreases the required reboiler duty (Figure 2)
2. decreases the bottom product temperature (Figure 3)
3. decreases the stabilized condensate TVP (Figure 4)
4. decreases the stabilized condensate H₂S content (Figure 5)
5. increases the required compressor power and cooler duty (Table 3)
6. decreases the stabilized condensate rate (Table 3)

To learn more about similar cases and how to minimize operational problems, we suggest attending our G4 (Gas Conditioning and Processing), G5 (Practical Computer Simulation Applications in Gas Processing), and PF4 (Oil Production and Processing Facilities), courses.

PetroSkills offers consulting expertise on this subject and many others. For more information about these services, visit our website at http://petroskills.com/consulting, or email us at consulting@PetroSkills.com.

By: Dr. Mahmood Moshfeghian

Reference:

1. Moshfeghian, M., January 2017 TOTM, PetroSkills | John M. Campbell, 2017.
2. Campbell, J.M., Gas Conditioning and Processing, Volume 2: The Equipment Modules, 9^th Edition, 2^nd Printing, Editors Hubbard, R. and Snow–McGregor, K., Campbell Petroleum Series, Norman, Oklahoma, 2014.
3. Soave, G., Chem. Eng. Sci. 27, 1197-1203, 1972.
4. ProMax 4.0, Bryan Research and Engineering, Inc., Bryan, Texas, 2016.

This tip is the follow up of December 2016 tip of the month (TOTM) [2] which investigated the benefits of having a water-draw and its optimum location in a non-refluxed condensate stabilizer column. That tip simulated the performance of an operating condensate stabilizer column equipped with a side water-draw tray to remove liquid water and a stripping sweet gas stream to reduce the H₂S content of a sour condensate. It determined and reported the possible locations of water-draw tray based water partial pressure along the column.

One option for stabilizer configuration is a split-feed design where a portion of the feed is pre-heated by heat exchange with the bottoms product. The remainder of the feed is fed to the top tray, similar to a standard “cold feed” stabilizer. Figure 1 presents an example of split-feed design. Split-feed provides heat recovery by pre-cooling the stabilized condensate upstream of the product cooler (not shown on this schematic), and reduces the required column reboiler duty (see also page 352 of Reference [1]).

Figure 1. A non-refluxed stabilizer column with feed-split, side water-draw and stripping gas

In order to lower the reboiler duty, a portion of the cold feed is heated by the hot stabilized condensate in a pre-heater. In addition, a free water knockout the upstream of the preheater is used to remove water and flashed gas from raw condensate. This tip will perform computer simulation to study the benefit of the upstream free water knockout drum and the pre-heater used in the split-feed design.

Specifically, this tip will determine if a liquid water draw tray is needed and how much reboiler duty is reduced by splitting the feed. This tip will consider stabilizing a sour condensate for Reid Vapor Pressure (RVP) specifications of 7, 7.5, and 8 psi (48, 52, 55 kPa). The tip will study the impact of the top feed-split % on the reboiler duty, bottom temperature, stabilized condensate H₂S content and True Vapor Pressure (TVP). The tip will present a summary of the computer simulation results and the key diagrams for the same plant.

Case Study:

Table 1 presents the raw condensate, containing about 21 mole % H₂S, and stripping gas compositions, rates and conditions. Figure 1 presents a simplified process flow diagram equipped with the preheater, stripping gas and water-draw tray for stabilization of this raw condensate. The 3-phase separator upstream of the preheater removes essentially all excess/free water. The tip utilizes   a feed splitter to heat up a portion of feed in the preheater by the hot stabilized condensate.

Table 1. Feed and stripping gas compositions, rates and conditions

The stripping sweet gas stream lowers the H₂S content of the stabilized oil and achieves the desired condensate vapor pressure at the specified reboiler temperature. For each case, the boil-up ratio was adjusted to meet the RVP specification. Table 2 presents the specified column variables.

Based on the data in Tables 1 and 2, and the process flow diagram of Figure 2, the tip performed simulation using the Soave-Redlich-Kwong (SRK) equation of state [3] in ProMax [4] software.

Table 2. Condensate stabilizer column specifications

Simulation Results:

The tip adjusted the boil-up ratio in the reboiler to meet the specified RVP of the stabilized condensate. The resulting boil-up ratio as a function of top feed-split is presented in Figure 2. This figure indicates that to achieve lower RVP at a specified boil-up temperature, higher boil-up ratio is required.

Table 3 also presents the summary of results for three specified RVPs of 7, 7.5, and 8 psi (48, 52, 55 kPa). This table indicates that decreasing the specified RVP decreases the stabilized condensate rate. Lower RVP requires higher vaporization of lighter compounds. The simulation results (not shown in this table) indicated that overhead vapor temperature of about 100 °F (37.8 °C) is practically independent of specified RVP and top-feed split. In addition, the results presented in Table 3 are independent of top-feed split which varied from 80 to 100 % with an increment of 2%.

Figure 2. Effect of top feed-split on the reboiler boil-up ratio for three RVP specifications

Table 3. Summary of simulation results for three specified RVP

Similarly, Figure 3 presents the required reboiler duty as a function of top feed-split and the specified RVP. This figure indicates that decreasing RVP increases reboiler duty and the reboiler duty increases linearly with the top feed-split %. The installation of the preheater reduced the required reboiler duty by about 18% for the case of top feed-split of 80%.

Figure 4 presents the variation of stabilized condensate temperature as a function of top feed-split. Decreasing the top feed-split from 100 to 80%, the bottom product temperature decreases by about 1.8, 2.1, and 2.4 °F (1, 1.2, and 1.3 °C) for RVP of 7, 7.5, and 8 psi (48, 52, 55 kPa), respectively.

Figure 5 presents the variation of stabilized condensate true vapor pressure (TVP) as a function of the top feed-split. Decreasing the top feed-split, increases the stabilized condensate TVP by about 2.7, 3.0 and 3.3 psi (19, 21, and 23 kPa) for RVP of 7, 7.5, and 8 psi (48, 52, 55 kPa), respectively.

Figure 3. Effect of top feed-split on reboiler duty for three RVP Specifications

Figure 4. Effect of top feed-split on bottom temperature for three RVP specifications

Figure 5. Effect of top feed-split on stabilized condensate TVP for three RVP specifications

Figure 6. Effect of top feed-split on stabilized condensate H2S content for three RVP specifications

Figure 6 presents the variation of H₂S content of the stabilized condensate as a function of top feed-split. This figure indicates that decreasing the top feed-split from 100 to 80%, the stabilized condensate H₂S content increases by about 56, 80, and 104 ppm for RVP of 7, 7.5, and 8 psi (48, 52, 55 kPa), respectively.

This figure also indicates that for the specified RVP of 7.5 and 8 psi (52, 55 kPa), the H₂S content exceeds the limit of 60 ppm at the top feed-split of 84 and 87 %, respectively. At the higher RVP, the bottom product temperature is cooler and there is not enough heat or stripping gas available to vaporize H₂S.

It should be noted that since the raw feed condensate leaves the feed tank (three phase separator) saturated with water but with no free water, simulation resuts showed no liquid water being trapped in the column. Therefore, no water was removed by the water-draw tray in all cases studied in this tip. This is contrary to the previous tip in which free water was allowed into the column and the traped water was removed by the water-draw tray.

Conclusions:

This tip investigated the impact of the feed-split on the performance of a non-refluxed stabilization column by varying the top feed-split from 80 to 100% by an increment of 2 % for three RVP specifications. Based on the simulation results, this tip presents the following observations:

1. Lower specified RVP, requires higher vaporization ratio at a given boil-up temperature. (Figure 2).
2. Decreasing the top feed-split from 100% to 80 %, decreases the required reboiler duty by about 18% (Figure 3).
3. Decreasing the top feed-split from 100% to 80%, decreases the bottom product temperature (Figure 4).
4. Decreasing the top feed-split, increases the stabilized condensate TVP (Figure 5).
5. Decreasing the top feed-split from 100% to 80%, increases the stabilized condensate H₂S content (Figure 6).

To learn more about similar cases and how to minimize operational problems, we suggest attending our G4 (Gas Conditioning and Processing), G5 (Practical Computer Simulation Applications in Gas Processing), and PF4 (Oil Production and Processing Facilities), courses.

By: Dr. Mahmood Moshfeghian

Reference:

1. Campbell, J.M., Gas Conditioning and Processing, Volume 2: The Equipment Modules, 9^th Edition, 2^nd Printing, Editors Hubbard, R. and Snow–McGregor, K., Campbell Petroleum Series, Norman, Oklahoma, 2014.
2. Moshfeghian, M., December 2016 TOTM, PetroSkills | John M. Campbell, 2016.

1. Soave, G., Chem. Eng. Sci. 27, 1197-1203, 1972.

1. ProMax 4.0, Bryan Research and Engineering, Inc., Bryan, Texas, 2016.