Onshore Gas Gathering Systems – Concept Selection, Basic Design & Operation (Part 2)
Onshore Gas Gathering Systems – Concept Selection, Basic Design & Operation
By: Mark Bothamley
Part 2
Increase/maintain the gas temperature above the hydrate temperature.
In this case, the gas will not be dehydrated but will be kept warm instead.
Although this is not a particularly commonly utilized hydrate prevention strategy, there are a few applications:
i. Low rate but high pressure gas wells with relatively short transport distance to a centralized processing facility.
ii. High rate, high pressure, high H2S wells.
In some cases, a choke/line heater is required at the wellsite to deal with the large JT expansion cooling effect experienced by choked high-pressure wells, especially during start-up. This is a somewhat different application than prevention of hydrates in the GGS but there are some common aspects to the equipment utilized.
First, the hydrate temperature of the flowing wellstream is estimated. From Figure 1, for 0.65 SG gas and assuming any free water present is condensed/fresh water, the estimated hydrate temp at an assumed average GGS pressure of 1,100 psig is ~ 65 F.
The next requirement is to estimate the flowing tubing temperature for the well. This is not a straightforward calculation and is one that most facilities engineers have little experience with. In the authors’ experience not many subsurface/production engineers have experience with this calculation either. It tends to “fall through the cracks” so to speak but is often important, especially for “near-wellhead” facilities work. The following equation provides an approximate method for estimating the flowing wellhead temperature for a given set of conditions:
Theoretically, this is an unsteady-state heat transfer calculation due to the conductance of the “infinite earth” surrounding the wellbore. There are also several heat transfer resistances – conduction and convection – involved in the buildup of the overall heat transfer coefficient, Uo, associated with the tubing string, tubing-casing annulus/annuli, cement layers, the earth itself (normally different materials that vary with depth), etc. Suffice to say, trying to calculate Uo from “first principles” is not easy. Instead, Uo values in the 2 – 2.5 Btu/hr-ft2-F range can typically be back-calculated from operating gas well data. The geothermal gradient varies regionally but a value of 0.0165 F/ft is typical.
Figure 5 below shows the results of typical wellbore heat loss calculations and corresponding flowing wellhead temperatures using the equation provided.
Figure 5 Flowing temperature profiles.
|
Bothamley |
This calculation provides an estimate of the flowing gas (and associated liquids) temperature at the wellhead upstream of the choke. If the well is being choked for flow control and there is a large pressure drop across the choke, there will be a Joule-Thomson expansion cooling effect across the choke. There are a number of approximate hand calculation methods – mostly charts, eg. Figure 6 – for estimating expansion temperature drop, but this is probably best done by a simulator. Clearly, low-rate, high pressure wells have the most hydrate risk. In fact, it is possible for these wells to hydrate off in the tubing below the surface of the ground, sometimes hundreds of feet below surface. These are typically high-pressure – though relatively low deliverability – wells, that are being choked to meet a low gas sales contract nomination. Flowing tubing pressures can easily be in the 1,500-2,000 psig range and higher. Combined with a low flowing temperature, the upper part of the tubing, wellhead and Christmas tree up to the choke are well inside the hydrate formation conditions region. Historically there have been a couple of ways to deal with this:
- 1. Inject methanol a. Down the tubing via a capillary string.
b. Down the tubing-casing annulus assuming the well doesn’t use a packer.
2. Install a bottomhole choke
a. this is basically a restriction nipple that is installed in the bottom of the tubing string to take the required pressure drop downhole where the temperature is warm rather than across the surface choke where it is much colder. This sounds good in theory but in the author’s experience, often did not work too well in practice.
Figure 6 PTH diagram - useful for wellsite choke heater design.
High flowrate gas wells can actually cause wellsite facilities/GGS problems by flowing too hot. What temperature is “too hot” ? There are several possible upper limit temperature constraints:
i) Max gas inlet temperature of ~ 120 F for glycol dehydration.
ii) Max temperature limit of ~ 140 F for flowline external coating (steel flowlines) or for the commonly used polyethylene layers of composite flowlines.
iii) Flowline thermal expansion/buckling issues.
iv) Increased volumes of condensed hydrocarbons and water in the GGS. The hydrocarbon liquids may not be a problem if they are going to be recombined with the gas anyway.
A typical wellsite choke heater is shown in Figure 7. These are glycol-water bath heaters that are often of split-coil design with a long-nose choke between the high and low pressure coils. A choke heater can be used to warm the gas from the wellhead thereby preventing hydrate formation associated with the expansion pressure drop. The temperature of the gas out of the low-pressure coil can be adjusted to improve wellsite separation/measurement or to heat the gas before putting it into the gathering system to keep the flowing gas warm on its way to the gas plant, as a hydrate prevention strategy. The bath liquid is typically a 50-50 mixture of water and ethylene glycol and operates at temperatures in the 180- 190 F range.
Figure 7 Typical wellsite choke heater.
Thermally insulated vs non-thermally insulated pipe.
Heat loss and the corresponding temperature profile for gas flow through a buried pipeline can be calculated from the following equation:
Area basis is pipe OD.
Figure 8 shows flowing temperature profiles for buried flowlines/pipelines for an assumed set of conditions and operational parameters.
Figure 8 Pipeline flowing temperature profiles.
Bothamley
Gas flow through non-thermally insulated pipe cools off quite quickly, especially for low flowrates. Transport of the gas over much more than several miles, will typically require the installation of multiple line heaters for periodic reheating of the gas in order to remain above hydrate formation temperature. While potentially feasible, this is not done very often. This type of system is very sensitive to flowrate, especially turndown operation.
Thermally insulated pipe improves the situation significantly. The main problem with this option is that insulated pipe is very expensive compared to non-insulated pipe. Another sometimes unappreciated drawback to this option, is that the pipeline is operating warm and wet, which for steel pipelines at least, results in a significant corrosion risk, even with no H2S and only 1-2 % CO2. In particular, top-of-line CO2 corrosion can be severe in downhill pipeline sections that predominantly operate in stratified flow. The author has personal experience with this problem which went undetected for years before rupture of the pipeline occurred.
Many high H2S “conventional” gas field developments utilize wellsite line heaters and insulated gathering system pipe to keep the wellstream above hydrate formation temperature all the way to the gas plant. Needless to say, materials selection and the corrosion control strategy utilized are major issues but can generally be dealt with.
iii) Hydrate Inhibition via Chemical Injection
For this option, the focus will be mainly on the use of methanol for hydrate inhibition. Ethylene glycol is also an option but is rarely used in onshore gas gathering systems, though perhaps it warrants additional consideration. Low-dosage hydrate inhibitors (LDHI’s) are also an option, though they are also rarely used for onshore applications. LDHI’s are expensive and the space and weight savings associated with their lower concentration requirements are not critical onshore.
Methanol is a so-called thermodynamic inhibitor which is soluble in water and depresses the hydrate formation temperature. Hydrate temperature depression is dependent on the concentration of methanol in the aqueous phase.
Methanol injection is typically the lowest CAPEX hydrate prevention option, often the main criteria for many companies, but of course has an ongoing chemical consumption operating cost. A significant consideration with the use of methanol is whether it is needed year-round or only in the colder winter months. As discussed previously, this mainly depends on geographical location (latitude). In the southern U.S. methanol injection in only the winter months is probably sufficient, while in northern Alberta, year-round injection – at least in high pressure systems – will be required.
The required methanol injection rate for a given application is dependent on several factors including:
i) The hydrate temperature depression required.
ii) The amount of liquid water present in the line.
iii) The amount of hydrocarbon liquid present in the line.
Methanol injection rates of 5 – 10 gal/MMSCF of gas are typical and can be determined more accurately by simulator or hand calculation. The methanol will distribute between the gas, water and hydrocarbon liquid phases. It is the methanol concentration in the water phase that is effective in preventing hydrates. The methanol volumes in the gas and hydrocarbon liquid phases are essentially “losses”, though the vapor phase methanol can be helpful in certain operational situations, eg. “melting” hydrate blockages.
Figure 9 Effect of methanol injection
An additional drawback associated with the methanol injection option – at least for steel flow lines – is internal corrosion. First, the system is being operated wet so the usual internal corrosion mechanisms (CO2, H2S, O2) are potentially active due to the presence of an electrolyte (water). Second, though not well – or widely – recognized, the presence of high methanol concentrations are known to actually increase corrosion rates under certain circumstances, especially in sour gas applications. Oxygen dissolved in the injected methanol is also a known contributor to internal corrosion.
In summary, as mentioned previously, methanol injection is probably the most commonly used hydrate inhibition strategy currently being used in most smaller flowrate, higher pressure, sweet gas gathering systems. In large flowrate systems, field located glycol dehydration units are more common. For high H2S gathering systems, wellsite – and intermediate if required – line heaters and insulated pipe are quite commonly utilized. Methanol injection provisions are nearly always provided as a backup to these alternative hydrate prevention strategies.
3. Sweet vs Sour Gas
Most onshore gas production in North America is sweet (< 4 ppmv H2S) but substantial volumes are sour (> 4 ppmv H2S) with H2S concentrations varying from 5-6 ppmv to 30+ %. Most tight/shale gas is sweet or at least < 100 ppmv H2S. The higher concentration (>1 %) H2S production tends to come from “conventional” fields. H2S removal in the field is generally avoided due to the cost, complexity, environmental and safety issues involved. There are exceptions, and these usually take the form of H2S scavenger systems, typically intend to remove ppm levels of H2S and less than a couple hundred pounds/day of sulfur equivalent.
From a field facility/gas gathering system point of view, low H2S gas can often be handled similarly to sweet gas, with extra attention paid to minimizing possible venting/emissions sources, with sweetening of the gas performed at a centralized processing facility. If the gathering system is operated by a separate company, they may accept, transport and process the gas with an additional “sweetening fee” charged, or may not accept the gas into their system, which would require the production company to install a scavenger sweetening system at the well/pad site.
Higher concentrations of H2S are not economically treatable with scavengers. This gas basically needs to be transported to a centralized gas plant – or possibly a large field compressor station equipped with an amine sweetening system (typically) – for processing and disposition of the recovered H2S/sulfur. As mentioned earlier, high H2S produced gas can be dehydrated at the well site/pad source to avoid hydrates and minimize internal corrosion during transport to the gas plant, or alternatively, well/pad site line heaters and insulated pipe can be used – for hydrate avoidance – along with corrosion inhibitor injection to protect carbon steel pipe. Some operators will choose to utilize corrosion resistant alloys (CRA’s) in these applications. They are expensive, but for large fields and long operating life, they will often result in life-cycle-cost advantages.
2. Liquids Handling
Most gas wells also produce varying amounts of hydrocarbon liquid and water that must be handled in some way. These liquid sources include:
1. Hydrocarbon liquids – mostly liquids that condense out of the gas due to temperature and pressure changes, but there may also be some free liquids that enter the well from the reservoir as well.
2. Free water – typically formation water and in early well life, flow-back water from the stimulation treatment. Produced formation water flowrates typically increase over time while flowback water production normally lasts for only a few months and is often separated at the well/pad site via temporary facilities. The specific aspects of frac flow-back water (and proppant) handling will not be discussed further in this article.
3. Condensed water – water vapor that condenses out of the gas phase mainly due to the drop in temperature of the gas as it moves up the well tubing and through the surface facilities.
For production accounting purposes, phase separation is typically required to allow measurement of the gas, hydrocarbon liquid and water (not all areas). Normally, “conventional” 2 or 3-phase separators are used. In some jurisdictions, individual wells are equipped with “wet gas” meters, basically orifice meters, with provisions – temporary or permanent – to periodically test the well with a conventional separator and meters. Multiphase meters are rarely used in onshore gas gathering systems. Figure 10 shows a typical conventional well site liquid handling arrangement.
Figure 10 Typical wellsite liquid handling options
The types and amounts of liquid that are present in the gathering system are mainly dependent on separation/processing decisions implemented on the well/pad site. These include:
Hydrocarbon Liquid Handling
Depending on wellstream composition and flowing temperature and pressure, some amount of hydrocarbon liquid - typically 0 – 50 BBL/MMSCF - will also be produced from the well. There are two main options for handling this hydrocarbon liquid:
i) Removal of free hydrocarbon liquids from the gas at the well/pad site.
The free hydrocarbon liquids are separated, measured and temporarily stored in onsite tanks then shipped out by truck – usually – or pipeline. These liquids are typically quite light and volatile, normally requiring some form of stabilization process – often simple flash separation – to reduce vapor pressure, in addition to water removal. If the liquids are sour, additional processing and precautions are required. Even with free liquids removal at the well/pad site, some amount of hydrocarbon liquid condensation can be expected in the GGS due to temperature reduction.
ii) Hydrocarbon liquids are recombined with the gas.
The free hydrocarbon liquids are separated, measured and then recombined with the gas for transportation via the GGS to the central gas plant for processing. Some amount of additional hydrocarbon liquid condensation can also be expected in the GGS due to temperature reduction.
A major factor here relates to the facilities ownership question.
Option (i) is commonly employed for “pad well” developments and/or for single well per wellsite developments where there is a change of ownership between the wells/pad site facilities and the GGS. In these cases it is common for the gas gathering company to have a “no free liquids” requirement in the gas gathering agreement which prohibits recombination of free hydrocarbon liquids with the gas for transport to a centralized gas plant.
If the operating company owns and operates everything from the wells to the gas plant, the hydrocarbon liquids are often recombined with the gas and “multi-phased” all the way to the plant for centralized handling and processing per Option (ii).
For many conventional gas field developments, Option (ii) is very common. While this option increases the liquid content in the gathering system, it has the following advantages:
i) Removes the need for hydrocarbon liquid storage at each wellsite.
ii) Removes the need for collection and trucking of hydrocarbon liquids.
iii) Eliminates flash vapor volumes, including potentially H2S, from the hydrocarbon liquid storage tanks.
iv) Transports the hydrocarbon liquids to a centralized gas plant where more efficient condensate stabilization and storage facilities are located.
v) Allows for a smaller wellsite footprint.
The main technical downside to recombining the hydrocarbon liquids with the dried gas is increased multi-phase flow related problems in the GGS, ie. higher pressure drops and slugging. Besides the potential multiphase flow issues, it will often be necessary, ie. for a wellsite glycol dehy system, to ensure that only minimal amounts of free water are entrained in the recombined hydrocarbon liquid. In rare occasions, asphaltenes and wax associated with the hydrocarbon liquids have also caused problems.
Shale gas wells are often equipped with packaged heater-separator units called Gas Production Units (GPU’s). See Figure 11. The GPU consists of two parts – 1) an indirect glycol/water bath choke heater and a separator. The heater section heats the well stream to prevent freezing when the high pressure fluids are expanded across the pressure letdown choke and is also used to reheat the lower pressure fluids to the desired separation temperature conditions. Some GPU’s are two-stage units – high pressure and low pressure – which helps with managing flash gas liberated from the hydrocarbon liquid condensate as it is reduced in pressure to storage tank conditions. Although GPU’s are commonly used on shale gas pads, they could also be used for conventional wells as required.
Figure 11 Gas well GPU’s on a shale well pad
While there are pro’s and con’s to each approach, these two hydrocarbon liquids handling options clearly have a large impact on gathering system line sizing and liquids handling.
Free Water Handling
Both of the dehydration methods outlined earlier remove water vapor from the gas stream.
There will normally also be free water in the wellstream, which comes from two sources:
i) Condensed water (fresh).
Potential volumes can be estimated with the use of a water content chart, shown earlier (Figure 2). For high sour gas, some adjustments need to be made to the calculated water content. Condensed water volumes of < 0.5 BBL/MMSCF are typical.
ii) Formation water.
iii) Depending on the reservoir characteristics and well completion details, some formation water will also likely be produced. Typical water-gas ratios are in the 2-10 BBL/MMSCF range, though higher produced water volumes are also possible. Flowback water volumes after hydraulic fracturing treatments can be very large and last for a significant period of time. Unlike condensed water, formation water is normally quite saline with overall total dissolved solids (TDS) typically in the range of 50,000 – 150,000 ppmw (seawater is ~ 35,000 ppmw).
For sweet gas wells, any free water produced from the well – condensed and/or formation water – would typically be separated from the wellstream, dumped to an on-site storage tank, and trucked out on a periodic basis. For high sour gas content wells, water handling is a bit trickier. Free water can still be dumped to a storage tank and trucked out but provisions must be made to control H2S in the vapor vented off the tanks. Very rarely, separated produced water is recombined with the gas and hydrocarbon liquid and all 3 phases are transported to the central gas plant for separation, treating and disposition. This has been done occasionally in high H2S developments (using line heaters and insulated pipe) to eliminate potential H2S releases from produced water storage tanks and truck loading/unloading operations. The corrosion/materials selection issues are even more difficult in this case.
Free water removal at the pad/well site combined with dehydration of the gas will result in a “water dry” gas gathering system which has significant benefits with respect to hydrate prevention and corrosion/materials selection.
Corrosion and Materials Selection
The main focus of this section will be on the flowlines/pipelines, which represent the primary components of a typical gas gathering system. Corrosion is essentially a metallic pipeline issue. Though non-metallic pipelines have potential degradation mechanisms as well, for the purposes of this article the discussion of corrosion is as relates to metallic pipelines. Both internal and external corrosion of metal pipelines need to be considered. External corrosion protection is relatively straightforward. The primary corrosive species is oxygen in moist soil, and the main protection measures are a good external coating supplemented with a cathodic protection system.
Internal corrosion control is more complicated and depends on several factors. Firstly, dehydrated systems – no free water – are usually free of internal corrosion, though there can be exceptions if occasional upsets introduce water into the GGS.
For wet systems (non-dehydrated), the main internal corrosion mechanisms are related to the presence of:
1. CO2
2. H2S
3. O2
These may be present individually or in combination.
Microbiologically induced corrosion (MIC) can occasionally be an internal corrosion issue but is fairly rare and will not be discussed further.
While there are numerous factors that impact potential internal corrosion severity, the most commonly employed material/corrosion control strategy employed for onshore GGS’s that choose to use metallic flowlines/pipelines, is carbon steel combined with a suitable corrosion inhibitor (CI). Corrosion resistant alloys (CRA) are rarely used – only for the most severe applications – due to their high cost. In fact, CRA’s are probably somewhat underutilized. Application of CRA’s is outside the scope of this article and will not be discussed further.
For mild to moderate corrosivity systems, ie. < 10 mpy uninhibited corrosion rate, carbon steel pipe, eg. API 5LX 42-52, combined with a nitrogen-based film-forming inhibitor is probably the most commonly employed “system”. Typical injection requirements for GGS applications are 1 – 2 pints/MMSCF. The actual protection effectiveness of the corrosion inhibitor depends on a number of factors, including, but not limited to:
1. Relative and absolute concentrations of CO2 & H2S.
2. Temperature and pressure.
3. Hydrocarbon liquid and water flowrates and compositions.
4. Flow pattern/phase velocity effects.
5. Inhibitor availability.
6. System cleanliness.
7. Pigging program employed, if any.
8. Etc.
Except for the most severe applications (where a CRA of some type should be used) a properly designed and operated corrosion inhibitor program can be very effective, ie. 90+ %, in mitigating internal corrosion of carbon steel pipelines. The primary requirement is to ensure that the polar CI molecules contact and adhere to the inside pipe metal surface. This can be especially difficult if there are solids/deposits – including corrosion products – in the system, or if the flow regime prevents contact of the inhibitor with the metal surface, ie. stratified flow with the inhibitor in the liquid and none in the vapor phase to protect the upper part of the pipe. This is particularly a problem in downhill runs of pipe where the flow pattern is nearly always stratified, and the fluids in the pipe are cooling down. This is a major contributor to so-called top-of-line corrosion. This situation often requires batch pigging treatments, with the corrosion inhibitor between two pigs, to ensure 360 degree contact of the inside pipe wall with the inhibitor.
Internal corrosion due to oxygen can be a major problem but is usually, restricted to very low pressure gathering systems. Many mature gas fields have GGS’s operating at low pressure – even sub-atmospheric – and oxygen ingress causes continuous problems, not just in the GGS but in the receiving gas plant as well. Many of the shale gas fields have vapor recovery units on their condensate tanks that inadvertently pull in air and cause difficulties with respect to the gathering system operators’ oxygen specification limit. Many 3rd party gas gathering companies will not accept gas from a facility that utilizes a storage tank VRU.