CASE STUDY: Low Cooling Water Flow

In this case study we have a chemical plant, operating with a two stage ejector system. The ejector system consists of one train of ejectors with a precondenser, intercondenser, and aftercondenser. The system has historically had poor vacuum and instability issues. The system is designed to be operating at a pressure of 88 mmHgA, however it was running closer to 123 mmHgA. This poor vacuum was having a negative impact on the process. A pressure and temperature survey of the system was performed multiple times under a yearly service agreement. The following was observed from the data.

All the condensers were piped in series, therefore the cooling water entered the precondenser, then went to the intercondenser, and lastly the aftercondenser. Because the condensers are piped in series, the cooling water flow only needs to be measured once for the entire system. A second cooling water reading could be taken at the outlet of the system to determine if any major cooling water leaks exist. The cooling water flow was measured to be 93 gpm, while the design is 200 gpm. The cooling water supply temperature was measured to be better than design at a temperature of 81 ºF, versus the design of 84 ºF. Since the cooling water flow was less than half the design, the cooling water outlet temperature was higher as would be expected. The precondenser cooling water outlet temperature was measured to be 98 ºF, versus the design of 92 ºF. Since the system is in series, the cooling water supply temperature to the intercondenser is the outlet flow from the precondenser. The cooling water supply temperature to the intercondenser is now slightly above design. The cooling water outlet temperature from the intercondenser was measured to be 114 ºF, versus the design 100 ºF. As with the intercondenser, the aftercondenser cooling water inlet temperature is now also elevated compared to the original design. This issues compounds itself as the cooling water moves through the system with each condenser seeing a subsequently higher cooling water inlet temperature when compared to each condenser’s design. The aftercondenser cooling water outlet temperature was measured to be 125 ºF, versus the design of 105 ºF.

Lower cooling water flow has a negative impact on the performance of each condenser. The tube side heat transfer rate is decreased due to lower cooling water velocity. In addition, the temperature rise of the cooling water is increased. Both of these effects combine to raise the pressure the condenser operates at. This puts additional back pressure on the upstream ejectors, forcing each ejector to compress out to a higher back pressure.  This often causes the ejectors to break critical operation, losing the internal supersonic shockwave, and causing a step change in the system’s operating pressure.

With the low cooling water flow and increased cooling water supply temperatures, the condensers were not operating as needed. The aftercondenser was operating at a pressure of 795 mmHgA versus the design of 785 mmHgA. Since the aftercondenser will set the discharge pressure of the second stage ejector, it was worse than design. The second stage ejector was therefore not operating correctly. The second stage ejector had a suction pressure of 220 mmHgA, which was worse than its design of 206 mmHgA. The same happened with the intercondenser having higher than design process side pressures due to low cooling water flow, which caused the first stage ejector’s discharge pressure to be worse than design. This caused a break in the ejector’s performance resulting in a poor process vacuum. Therefore, the system was not operating within design throughout the entire vacuum system.

A study was done to evaluate ways to help increase the cooling water flow and bring it closer to the 200 gpm design.  The resulting recommendation of this study suggested a booster pump should be added to increase the cooling water flow to 150 gpm. Another pressure and temperature survey was completed and the vacuum system was operating at 100 mmHgA, versus the design of 88 mmHgA. This was a marked improvement over the previous pressure reading of 123 mmHgA, but not quite where the vacuum system should be operating. It was found that the cooling water flow was being throttled back at one of the cooling water isolate valves. Once the valve was opened completely, the system was able to get the full cooling water flow rate. An additional pressure and temperature survey was completed and the cooling water flow was measured to be 205 gpm, versus the design of 200 gpm. With the cooling water flow slightly better than design, the operating pressure was measured to be 86 mmHgA versus the design of 88 mmHgA. The vacuum was now in line with the system’s design.