Strange phenomenon of internal leakage in the drain line of the ammonia discharge pipeline in the condenser tower

I would like to ask the original poster: Is it permissible to install a backwash system on the high-pressure side of the equipment? Is the valve leak test conducted under the same pressure conditions? If that’s the case and leakage still occurs, it might be due to: 1. Presence of debris at the valve’s sealing surface; 2. The sealing surface being exposed to cold, which causes thermal stress and leads to damage to the seal: lol

We are a medium-sized nitrogen fertilizer plant that uses traditional manufacturing processes. The short service life of the ammonia discharge self-regulating valves in our condensation towers has been a persistent problem for us; this is also a challenge that is difficult to overcome in the nitrogen fertilizer industry. We once had to shut down production altogether because of severe internal leakage in these self-regulating valves. To address this issue, we installed drain valves in addition to the self-regulating valves, so that when there is significant leakage, it is possible to relieve pressure and replace the valves without shutting down the plant.

New Advances in Synthetic Ammonia Release Valves Abstract: The service life of the new type of valves designed for automatic ammonia release has exceeded one and a half years, effectively solving the difficulties associated with level control of ammonia in synthetic ammonia production as well as in heat exchangers. It introduces the research progress of ammonia release valves, as well as their new structure, working principle, and application scenarios. Keywords: ammonia release valve ； two-phase flow ； valve core ； sleeve valve ； Caliber 1: Introduction. Both the cold exchanger and the ammonia separation unit in the synthetic ammonia plant require control of the ammonia liquid level, as the liquid ammonia layer serves to prevent gases from the high-pressure system from entering the low-pressure system. Generally speaking, this is a straightforward ordinary level control project that is easy to implement. However, the extremely short service life of the ammonia release valve (it loses control after about a month) prevents this simple level control from functioning. Over the past 40 years, the fertilizer industry, manufacturers of instruments and valves, as well as research institutions have done considerable work to develop various control valves for handling high-pressure differential fluid flows. The actual effects vary, and the lifespan is mostly within half a year. It represents a significant improvement over conventional valves, but it does not solve the fundamental problem. Starting from the failure mechanism of the ammonia release valve, I studied various other types of ammonia release valves and, by taking comprehensive measures in terms of structure and materials, developed a new type of valve specifically designed for ammonia release. Currently, in the applications at Xiangdong Fertilizer Factory in Youxian County, Hunan, and the Fertilizer Factory of Fujian Yong’an Zhisheng Chemical Joint Company, the longest service life has exceeded one and a half years. Judging from the valve’s opening degree, it still has a long service life ahead. 2. Damage mechanism of the ammonia release valve 2.1 The presence of two-phase fluid is a specific physical phenomenon in the ammonia release process. Two-phase fluid arises due to two completely different factors. One method is that when liquid ammonia under high pressure passes through a small gap in a valve, its static pressure drops below the saturation vapor pressure, resulting in flashing. However, due to the short throttling zone, vaporization is incomplete. It exhibits a two-phase liquid coexistence state, in which the volume of the gas exceeds that of the liquid ； Another reason is that liquid ammonia produced under high pressure contains a certain amount of hydrogen and nitrogen gases dissolved in it, and the significant pressure reduction during the ammonia release process causes these hydrogen and nitrogen gases to come out of the liquid ammonia. According to relevant information, when 1 m³ of liquid ammonia is depressurized, 48 m³ of hydrogen and nitrogen gas are released. It is clear that the volume of the gas in the resulting two-phase fluid **exceeds the volume of the liquid**. Therefore, the droplets of unvaporized liquid ammonia are suspended in the vaporized ammonia and hydrogen-nitrogen gases, and move forward surrounded by them. 2.2 The impact of high-speed ammonia droplet streams causes excessive wear on the valve core. For a single gas, its molecular mass is low, allowing for very high flow speeds. A single liquid, with its low molecular mass, can have a very high flow velocity. A single liquid has a high molecular mass, but its flow velocity cannot be high. Therefore, a single fluid cannot exert destructive force on the valve core. In a two-phase fluid, the liquid ammonia droplets suspended in the gas possess the mass of a liquid, as well as the velocity of a gas. Mass multiplied by velocity results in kinetic energy that is far greater than that of a single gas or liquid, thereby exerting an exceptional destructive force on the valve core and seat. This damage process is similar to the sandblasting process, in which air acts as a carrier that moves abrasives forward, with the speed of the abrasives being close to that of the carrier. During the flashing process, the liquid droplets act as abrasives, while the carrier is hydrogen, nitrogen, and gaseous ammonia. Their motion state is exactly the same; although their destructive power is slightly lower, severe cases can still damage a valve within a period of one week to one month. On-site observation showed that the damaged valve core had a very smooth surface, as opposed to the honeycomb-like rough surface caused by cavitation; this indicates that both chemical corrosion and cavitation effects were minimal. The main cause of the damage is the physical destruction caused by the two-phase flow generated by flashing. Gas dissociation during the depressurization process exacerbated this damage. And it ensures that the two-phase flow exists irreversibly throughout the entire ammonia release process. 3. Structural features of various ammonia release valves and measures to address lifespan issues. Over the years, in order to solve the problem of the lifespan of valves operating under high pressure differences, valve manufacturers have produced various types of valves, and end-users have also made many attempts at improvements for use in ammonia release applications. The main measures are essentially just a few: that is, reducing the voltage in stages ； Change the liquid flow flicking method ； Offset the sealing surface from the worn surface ； High-strength materials are used. Practice shows that some measures are ineffective or only have a minimal impact, and it is also evident that relying on a single measure is far less effective than using a combination of multiple measures. The key to the problem lies in whether it is possible to address the damage mechanism of the valve effectively. The following discusses several typical structures. 3.1 Pressure reduction via multi-stage spools As shown in Figure 1, this control valve utilizes a multi-stage spool structure. The valve core and seat work together to divide the overall pressure difference of the control valve into several smaller pressure differences, reducing the pressure step by step. The design intent is to ensure that each differential pressure does not exceed the critical pressure. The purpose is to prevent the formation of a two-phase flow due to the generation of gas from the flash evaporation of liquid ammonia when the critical pressure is exceeded. This structure and method are somewhat effective, but they cannot solve the fundamental problem. The reason is that the gas generated by desorption during the ammonia release process accounts for a large proportion; desorption does not cease just because the pressure is reduced below the critical pressure. Therefore, even if flashing is avoided, two-phase flow cannot be avoided. At best, it only reduces a small amount of flash gas, slightly alleviating the contradiction. The method of adding throttling elements before and after a valve, which is similar to the co-phase method, has roughly the same principle and effects. 3.2 Special valves with channel-type spools: The design approach for these valves produced by a factory in the southwest is still based on the method of distributing pressure drops. It involves lengthening the throttling area between the spool and the valve seat as much as possible, thereby creating a pressure-reduction channel that allows the pressure drop to be distributed along the length of this channel. Secondly, since the taper of the valve core can be made very small, the angle between the flow direction and the axial direction of the valve core is also small. As a result, the component force exerted by the fluid ejected from the throttle hole inlet perpendicular to the surface of the valve spool naturally decreases. Furthermore, the hardness of the valve core material was increased after treatment. Therefore, this type of valve has become one of the better-valued valves in use today. However, for the same reason: the pressure drop due to dispersion cannot prevent two-phase flow, so the fundamental problem remains unsolved. The typical service life is about half a year. At the same time, the increased hardness of the valve core has introduced new problems. The diameter of ammonia valves is generally very small; when long ducts are used, the diameter of the valve core becomes even smaller, and it becomes fragile due to increased hardness. Moreover, the structure of long ducts causes large radial shear forces to arise when the valve stem slides into the valve seat if there is a misalignment; even a slight deviation in the alignment between the valve stem and the valve seat can lead to the breakage of the valve stem. This situation is quite common. 3.3 Structure to keep the sealing surface away from the impact of two-phase flow: A reverse-cone-shaped valve core, designed with the aim of minimizing the angle of attack of the component in the direction of fluid flow, thereby preventing the fluid from striking the component surface perpendicularly. At the same time, offset the worn area from the sealing surface. In fact, this structure is more suitable for manual valves. Automatic adjustment makes it difficult to meet the requirements when an equal percentage flow characteristic is needed. In use, the lifespan of this type of valve is also short. 3.4 Hard surface surfacing of valve cores and seats or use of new materials: Stellite alloy (Rc45) and hardened tool steel (Rc70) are widely used both domestically and internationally to manufacture valve cores and seats that are resistant to cavitation. Some of these materials are very hard and also very brittle. To achieve better overall mechanical properties, these materials are cladded or sprayed onto certain stainless steel substrates (see Figure 4). Generally, this method is suitable for valve spools and seats of larger sizes. Most of the ammonia synthesis plants in our country have an annual production capacity of less than 100,000 tons; the diameter of the ammonia release valves is very small, and their valve cores are also small, making them difficult to manufacture. At the same time, it has a small diameter and high material hardness, making it easy to break. The author has previously used pre-coated zirconium materials to manufacture valve cores and seats; on the one hand, the cost is high, and on the other hand, the results are not entirely satisfactory. When the oxide film on large-size valve cores is well controlled, their service life is relatively long. It’s very difficult for the small valve core to last more than half a year. In fact, all various improvement solutions involve improvements to the ingredients used in the diet. However, by considering only the materials without changing the structure, it is impossible to significantly increase the service life. 3.5 Main improvement measures for the new ammonia release valve 3.5.1 Adoption of a sleeve valve structure The sleeve-type structure has long been one of the basic structures for control valves, but currently there are no such products available in China’s high-pressure valve series. Even in the series of sleeve-type low-pressure valves, there are no products in the small-diameter sizes. The use of a sleeve structure in valves with high pressure differences and low flow rates is an innovation. Based on such considerations, the designer ensured that the fluid injected through the symmetric small holes in the sleeve wall became a two-phase fluid; on one hand, the fluids collided with each other at the center of the sleeve, creating intense disturbances. These numerous disturbances caused the vast majority of high-speed droplets to lose their kinetic energy and destructive power, with that energy being converted into heat energy due to the frictional resistance of the fluid. On the other hand, the ejected droplets reach high speeds only at the center of the sleeve, where they gain destructive power. At this point, even those droplets that can pass through the center and those that still possess a secondary velocity after impact are hindered from reaching the opposite wall of the cylinder by the liquid remaining in the cylinder cavity. The liquid provides self-buffering energy dissipation. This prevents direct damage to the valve plug (equivalent to the valve core) and the cylinder wall, thus protecting the sealing surface. Another structural advantage is that the valve plug always moves up and down inside the sleeve, with the sleeve serving as a guide. There is no phenomenon in ordinary angle valves where, during opening and closing, the valve core breaks due to radial shear forces on it caused by the misalignment between the valve core and the valve seat. It is clear that the use of a sleeve-type structure can solve many problems that other structural forms cannot address. However, it is not perfect. The valve plug moves within the sleeve, and there must always be a gap between them; as a result, leakage is inevitable. To completely eliminate leakage, the structure would become very complex. Fortunately, the ammonia-release process involves a continuous flow of fluid over an extended period, so leaks are not a concern; what matters is whether the leaks are small enough. When the leakage is reduced to the minimum level of ammonia production, the liquid level can rise again after the valve plug is closed, and the leakage rate remains constant over time, which meets the production requirements. 3.5.2 Use of a “side-in, bottom-out” flow direction: Generally, using a “side-in, bottom-out” approach for high-pressure angle valves can reduce wear on the valve core. This approach is adopted here to achieve such a purpose. Due to the \"side-in, bottom-out\" design, the bottom surface of the valve plug is in a low-pressure area; after the unbalanced forces are addressed, the upward thrust on the valve plug is minimal, and the operating force required for the diaphragm head can also be reduced. More importantly, the area above the valve plug is also a low-pressure zone; the high-pressure packing only needs to provide low-pressure sealing, which makes it easier to resolve sealing issues. 3.5.3 Use of high-strength, high-hardness alloy materials: Although the droplets in a two-phase fluid possess significant impact kinetic energy, they are still liquids and lack the mass and hardness of solids. As long as the internal components of the valve, particularly the throttle element such as the valve sleeve and valve plug, are made of high-strength and high-hardness materials ； As long as the strength and hardness of these materials are high enough, they can withstand the damage caused by two-phase flow. New types of valves have found and adopted this material. 4. Some issues in the application of new ammonia release valves 4.1 Selection of the diameter and flow capacity of ammonia release valves The selection of ammonia release valves is essentially a routine design process, with no new issues to discuss. However, issues related to diameter and flow capacity often become points of contention between designers and project owners. Therefore, it is specifically clarified here. Based on an annual ammonia production capacity of 10,000 t/a, the calculated values are as follows: Q1 = 2.48 m3/η (flow rate), ρ = 0.58 g/cm3 (density), P1 = 29,000 kPa (pressure before the valve), p2 = 1,600 kPa (pressure after the valve), t1 = 30°C (liquid temperature before the valve), D1 = 1,600 kpf (pipe diameter), Fl = 0.8 (pressure recovery coefficient), PC = 11,378 kPa (thermodynamic critical pressure), Pv = 1,216 kPa (saturation vapor pressure at the valve inlet temperature). According to the III J-type process configuration, 75% of the ammonia is separated in the cold exchanger, while the remaining portion is separated by the ammonia separator; thus, for the cold exchanger, Kv = 0.145 × 0.75 = 0.108, and for the ammonia separator, Kv = 0.145 × 0.25 = 0.036. Using these calculation results, it is possible to easily determine the flow capacity required for the valves in plants with various production capacities. The maximum flow capacity of the Dn15 valve can reach 4.0, which can meet the production requirements of any manufacturer in China. Therefore, there is no need at all to choose a larger caliber. 4.2 Configuration of valve pipelines and requirements for valve capacity control: As shown in Figure 6, the ammonia release valve is piped using a side-in, bottom-out configuration. The diameter of all pipes and associated valves is Dn15. Because there are catalyst particles and debris in the pipes within the synthesis section, and the orifices of the control valves are small, they become clogged very easily. Therefore, a dedicated high-pressure filter is installed. Under normal operating conditions, the two shut-off valves before and after the control valve must be fully open to ensure that all pressure is applied to the control valve. There is almost no pressure drop across the globe valve, so it is less likely to get damaged. Once maintenance is required, it can be shut down without the control valve becoming inaccessible due to leakage. The shut-off valve on the bypass must be completely closed, as ordinary shut-off valves do not have any design features to prevent cavitation, nor are they reinforced in terms of material. The bypass shut-off valve will become damaged quickly after use, and leakage cannot be avoided even if it is closed again. At this point, even when the level gauge indicates the lower limit, the drop in liquid level cannot be controlled; this often leads people to think that there is a leak in the control valve, but that is not the case. Of course, the fundamental solution is to design a new stop valve based on the principle of the ammonia release control valve. 4.3 Common fault phenomena and their solutions 4.3.1 Clogging and jamming: Due to the oil contamination generated by the synthesis system, as well as desorbent particles and powder, welding slag, and other metal debris, filters are installed. However, the capacity of the filter is limited; therefore, the maximum pore size through which solids can pass through the filter is set at Ф2mm. A large amount of particles and powders smaller than Ф2mm that can pass through the control valve are allowed to flow through, with only objects larger than Ф2mm being intercepted. However, those individual small particles whose dimension in one direction is less than Ф2mm while their dimensions in other directions are greater than Ф2mm (most typically amorphous catalyst particles) may pass through the filter. These small particles may get stuck in the flow holes of the sleeve and extend into the sleeve, preventing the valve plug from closing. Although this situation occurs very rarely, once it happens, the use of the stop valve must be halted, and it needs to be disassembled and cleaned in order to resolve the issue. Therefore, more efforts should be made in terms of ammonia distribution and cooling exchanges to prevent contaminants from passing through the pipeline connected to the automatic control valve as much as possible. 4.3.2 Self-oscillation: Self-oscillation occurs in the control system loop, as well as in the small loop formed by the control valve and the valve positioner. Any type of self-oscillation will cause severe wear on the valve packing, so it must be avoided. The reasons for the former situation are more complex; it depends on which part of the system is problematic, and the issue must be addressed accordingly. I encountered this situation at the Yueyang Fertilizer Factory; due to improper handling, it led to premature wear of the packing. Ammonia leakage from the filler occurred after about two months, forcing us to stop using it. In fact, the valve core is still intact. Self-oscillation can also occur in the small closed loop formed by the control valve and the valve positioner. The reason is that a phase shift occurs at each stage of the forward path of this closed loop. When the diaphragm head of the valve is small, the resistance and capacitance correction elements present in the system may not be able to overcome enough phase shift, causing negative feedback to turn into positive feedback and thus resulting in self-oscillation. For this reason, the design suggests that small diaphragm valves are not suitable to be used in conjunction with electrical valve positioners (especially type П), and that electrical converters should be employed instead. 5. Conclusion One approach is to prevent the formation of two-phase flow, which is logically sound. However, there are two lakes with two-phase flow, but only one was taken into consideration. Therefore, trying to solve the problem fundamentally fails to resolve it at all ； Another approach is to acknowledge that two-phase flow is inevitable, and then take advantage of the properties of the fluid to implement measures to avoid its consequences. The effect is quite noticeable. The significant extension of the service life of the ammonia release valve is achieved through the correct diagnosis of the causes of damage. The defensive measures adopted are comprehensive and fairly thorough, thereby enabling good results to be achieved

Severe internal leakage in the ammonia discharge valve can be caused by two factors: 1. Prolonged air leakage into the separator leads to severe erosion of the control valve’s valve seat (even slight air leakage can cause this problem)； 2. The material of the control valve itself is of poor quality; it is a problem related to the quality of the control valve. I’m not sure what the reason is on the part of the poster, but by taking appropriate measures, this issue can be significantly alleviated; at least it won’t cause the system to stop working.