Isotherm, Inc.

Is Shell and Plate Heat Exchanger a better choice for Ammonia/Carbon Dioxide Condenser?

There is no ideal product for a certain application. Each product type has its pros and cons; however, as engineers it is our job to select suitable equipment with close attention to risk analysis. In the case of Ammonia/Carbon Dioxide Condenser, besides the space and economic aspects, it is important to pay close attention to the mechanical integrity of a system.

In a CO2/NH3 cascade system one of the most important components is the cascade condenser. Any potential leak would have a devastating effect on the operation of the system. It is a well-known fact that mixing of ammonia and carbon dioxide results in an ammomium carbomate powder that would rapidly contaminate the entire system and play havoc on the plant, resulting in replacement of condenser, compressor, and other auxiliary components such as piping and control valves.

Since the introduction of Shell and Plate heat exchangers, contractors have been incorporating them as oil coolers, evaporators, water cooled condensers etc. This trend has also been extended to Ammonia/Carbon Dioxide heat exchangers because of their efficiency and compactness; however, ignoring the fact that these heat exchangers could be vulnerable to material and/or weld failures simply because of the construction involved in the making of them. Figure 1 shows a typical Shell and Plate Heat Exchanger. Figure 2 shows a plate pack prior to insertion in a shell and Fig. 3 shows a close-up view of adjacent plates being fused together.

It is important to note that with hundreds of “Fused Only” welds, the risk of small pinhole or/and lack of fusion could end up in a catastrophic failure during operation. The weak weld/s may not compromise during factory testing, though it could after several pressure and thermal cycles during operation. Historically it has been found that the cause of failure was always due to fatigue stress factor. Figures 4 and 5 show a bad weld and a good weld at the periphery between the adjoining plates of a failed heat exchanger. A weak joint such as shown in Fig. 4 did not fail at the time of the pressure test in the factory; however, after several cyclic operations the crack propagated and resulted in a major leak.

Fig. 1 Typical Shell and Plate Heat Exchanger

Fig. 2 Plate Pack

Fig. 3 Adjacent plates being fused together

Fig. 4 Failed weld

Fig. 5 Good weld