In this case study we have a vacuum distillation tower, operating with a three stage ejector system. The ejector system consists of two trains of ejectors, train A and train B, that are run in parallel. The system has historically had poor vacuum and instability issues. The tower top is designed to be operating at 11 mmHgA, however it was running closer to 22 mmHgA. This poor vacuum was having a negative impact on the performance of the vacuum tower. A pressure and temperature survey of the system was performed multiple times under a yearly service agreement. The following was observed from the data.
The system was operating with only the A train of ejectors in service. When initially surveyed, the motive steam pressure was measured at the header one deck below the ejectors, because that was the only functioning test location. The pressure was found to be 135 psig, which is better than the design. Often the motive steam pressure is different at various ejectors in a system due to piping pressure losses. Therefore, the pressure should be measured at each ejector to ensure the steam is adequate to all ejectors. In a vacuum distillation system, the lowest motive pressure is often identified at the first stage ejectors, as they normally are installed at the top of the system which is at the end of the steam header.
The site had purposely throttled back on the motive steam to the operating first stage ejector. This had the effect of making the performance stable but caused the first stage ejector to operate in a broken manner without the critical internal shockwave. When ejectors operate in this manner they are stable but with a reduced compression, which results in a poor suction pressure relative to the design. With only one train of ejectors in operation and the first stage ejector throttled back on steam pressure, it had the effect of reducing the steam consumption of the system. The lower steam consumption allowed for less pressure loss in the steam lines, which resulted in better steam pressure to the downstream second and third stage ejectors, allowing them to remain stable and operate with their internal shockwaves. While these operational changes were helpful in creating stability and achieving the best possible operation with the poor steam conditions, this did not fix the underlying reason for the poor performance. In order to fully correct the issue and achieve a good vacuum, the underlying poor motive steam pressure needed to be addressed.
In order to correct low motive steam pressure, the following items should be reviewed:
- Verify that all valves in the piping are open and all strainers are blown down and clear so that no unnecessary pressure drop exists. Insure the valves are in good working condition.
- Adjust the let-down station or increase motive steam header pressure to increase the motive pressure at the ejector’s inlet.
- Redesign the motive steam nozzles and replace them to allow them to work at a lower motive steam pressure.
Additional testing was done on the B train ejectors that was not in service. The steam pressure at the header was measured to be 145 psig with only the A train in service. When the B train’s third stage ejector was put into operation, the motive pressure dropped to 139 psig. When the B train’s second stage ejector was put into operation, the motive pressure dropped further to 135 psig. When the B train’s first stage ejector was put into operation, the motive pressure dropped even further to 128 psig. This is lower than the design pressure of 130 psig. The motive steam pressure to the header was maxed out for this test and higher than it normally ran. Additional steam pressure was not available to the header. Achieving the design motive pressure to the ejectors and reliably maintaining it was not feasible with both trains of ejectors in operation.
The solution to the motive steam issue was to have Graham redesign the motive steam nozzles so that the system would function properly at the lower motive steam pressure. The internal geometry of the nozzles was adjusted so that the same amount of energy was passed into the ejector, but at a lower motive steam supply pressure of 120 psig. This would allow a lower motive pressure to be supplied to the ejectors while allowing all of the ejectors to operate in a critical manner. Being able to utilize the second train of ejectors and have the first stage ejectors operate in a critical manner would result in an improved tower top pressure.
During an outage, the new nozzles were installed in every ejector in the system. When the system was brought back online, the motive pressure at the first stage ejectors was at the 120 psig design, in accordance with the new nozzles. This resulted in an improvement in the tower top pressure to 11 mmHgA, in line with the system’s design. A marked improvement over the previous operating pressure of 22 mmHgA. Subsequent follow up visits confirmed the continued operation of the system near the design tower top pressure.
This case study shows the importance of the motive steam pressure in the successful operation of an ejector system. Knowing the motive steam pressure at the inlet of each ejector and maintaining that pressure within the range of -0% / +20% will help ensure good performance. While there may be things that can be done to boost a low steam pressure up to the design, if that is not possible new redesigned motive steam nozzles may be a good solution.
