Over the years, one of Jake’s most difficult troubleshooting exercises had been diagnosing a mechanical failure in a heat exchanger. Many symptoms can be attributed to such performance-robbing phenomena. Mechanical failure can require shutting down and opening the heat exchanger to look for the cause. Jake knew many young engineers whose careers were ruined after rushing to an expensive shutdown to solve a problem only to find it revealed nothing. So, he took the equipment offline only after exhausting every avenue.
[pullquote]
Troubling Flow
Charlie called Jake to the plant to assess a problem with a very low temperature evaporator. Performance suddenly had dropped. The plant had difficulty keeping the unit online but the site needed all machines to meet demand.
Charlie told Jake that the Saturday midnight shift became aware of the problem when the evaporator pressure rapidly decreased. The chiller struggled to maintain the very low temperature setting. The flow had spiked but then settled down. Charlie showed Saturday’s chart of the flow. The flow did indeed spike but returned to design within minutes. The evaporator then showed a quick reduction in pressure but stabilized significantly below design. This was one of four cascade systems that supplied a secondary refrigerant to the process area. The blended flow and temperatures kept the process requirements within limits but the supply temperature was coming close to the maximum allowed.
Charlie went through the checks they had performed. They first looked at the instrumentation. Most important was the flow measurement and that was within specification tolerances. Everything else was right on specification. A heat transfer analysis of the evaporator showed the temperature differences had spiked following the flow event. The resultant evaporator pressure was running near the minimum the system could produce. The temperature differential across the secondary coolant was decreasing gradually.
“So, Charlie, what do you think is wrong?” Jake asked.
Charlie replied, “I think we have to take the evaporator offline.”
“And why is that?” Jake inquired.
“Because we have a blinded tube sheet!” Charlie blurted out.
“Okay, let’s review the logic,” Jake replied. “You may be right, but we need to make sure as I’m not convinced. You and I both know the importance of having this unit online right now.”
They reviewed what they knew and the logic behind the hypothesis. The flow spiked, then the flow controller returned it to set point. Following the spike, the chiller attempted to correct the outlet temperature but the log mean temperature analysis showed the compressors’ capability to produce the lift required restricted the chiller. The outlet temperature increased as it reached the limit. In the past, residue in the secondary refrigerant system had broken loose and blinded the tube sheet. A momentary blockage of the tube sheet could have caused the spike in flow could but that would have resulted in a higher pressure drop across the tube sheet after the flow was corrected. That did not happen.
“So, Jake, what do you think it could be?” asked Charlie.
“Well I have seen something like this before. It might be a damaged baffle between the passes. We will still have to take the chiller down to find out,” replied Jake.
The flow’s spiked increase indicated to Jake that the flow restriction had been removed. The two-pass heat exchanger’s flow spike was large. After the flow returned to set point, the pressure drop was slightly lower than design, but that could be misleading. It did seem to indicate some major physical event, meaning they’d need to take the chiller offline.
Once off line, Jake and Charlie found the seal had cracked and blown out, allowing flow to bypass the tubes. The mixing of flow through the tubes with the bypass flow meant the chiller outlet temperature would increase due to the chiller’s limited capability.
