Imagine if there were no way to dissipate heat in manufacturing processes or food processing. The modern conveniences that we enjoy today would not exist.

Chillers are as essential to industrial products as the process equipment they cool. They protect valuable investments, products and plant infrastructure while keeping processes running at optimum efficiency.

Whether used to cool a process critical to delivering product on time to a major customer, or simply to avoid the costs and disruption resulting from equipment failure, a chiller delivers. Many options exist when selecting a chiller for your process-specific needs. Matching process expectations with the appropriate chiller design will help achieve the highest level of availability, commonly referred to as uptime.

Improve Chiller System Reliability

Reliability describes the ability of a system or component to function under stated conditions for a specified period. Intuitively, it is easy to understand that a product’s design, the quality of materials used and the care taken during the manufacturing process all influence the quality and reliability of the products we select.

Chillers are no different. If low cost components designed for relatively short lifespans are selected, then the reliability of the overall system will reflect those choices.

Reliability engineering is an involved science and beyond the scope of this article. However, there are a couple of basic concepts that are useful to understand: mean time between failures (MTBF) and mean time to repair (MTTR).

  • MTBF is the reciprocal of failure rate, and it is an estimate based upon averages usually derived from testing or field experience.
  • MTTR is the time it takes to diagnose a failure, contact a repair technician and to restore the system to operation.

To convey value, many manufacturers will determine and publish MTBF data for a component product such as an electric motor. This information is used by design engineers to select components appropriate to meeting product design goals.

Using MTBF and MTTR, you can determine the expected uptime or availability for a given system. Availability is expressed as a ratio of MTBF over the sum of MTBF and MTTR. As figure 1 shows, the higher a system’s MTBF, the higher its availability. Conversely, the longer it takes to repair, the lower the system’s availability.

MTBF can be improved in three ways:

  • Select high reliability components for application under conditions for which the component is intended.
  • Add redundant components or subsystems to back up lower reliability elements.
  • Use modular redundant systems to provide backup capacity in case of system failure.

The impact of each of these options can be visualized using a simple model with the help of a chiller system block diagram. As shown in figure 2, all components must function for the system to function, equating to a serial system in reliability parlance. The reliability of a series system will always be lower than that of the least reliable component, despite the common assumption that a system’s MTBF is equal to the lowest MTBF component. The equation in figure 2a shows how using high reliability components will increase system reliability.

Add Redundancy to Extend Chiller Life

Redundant components or subsystems can further improve system reliability by shifting some or all of the system design from a purely series design (as discussed above) to a fully or partially redundant, parallel design. A fully redundant design can be achieved with two identical chillers, each capable of providing the cooling capacity required. The downside of this approach is essentially a doubling of cost, yet it can be appropriate for some highly critical applications.

Another approach would be to look at the task from a component and subsystem level. With this approach the objective is to add redundant elements where failure is most likely.

Continuing the example shown in the figures, it is evident that the pump is the least reliable component in the serial chain. In cases where more reliable components are either unavailable or cost prohibitive, the system’s MTBF can be improved by adding a backup component. The resulting system MTBF is calculated as shown in figure 3. In this case, adding a redundant pump improved the system MTBF by 50 percent.

The concept demonstrated by the example can be applied to other components such as compressors. Unfortunately, some rare failure modes still can compromise remaining compressors and other system components. Real-world chiller systems with redundant, isolated refrigeration subsystems are available. They may be an appropriate solution if the impact of process downtime warrants the increased cost of higher reliability.

Allow for Growth with Modular Systems, Assess Repair Time

Redundant modular systems often are selected for larger central systems because they offer the ability to scale up as an operation grows. At the same time, they provide the ability to service individual modules while others remain operational. The redundancy provided by these systems is often referred to using the concept of N+1, where N represents the largest chiller required to meet the process need. The +1 refers to the situation where the second module provides full backup capabilities in the event of module failure.

Table 1 explains several options for configuring the number and capacity of each module to provide the needed cooling capacity. The excess installed capacity resulting from each choice and the remaining capacity after one module failure are shown.

The 2/3(N+1) example, for instance, take into account worst-case conditions such as maximum process load and, for air-cooled systems, high ambient conditions. Robust design combined with the concept of 2/3(N+1) can provide a level of redundancy under normal (partial-load) operating conditions. As the example shows, dividing the load between smaller chiller modules results in at least some operational capacity when equipment fails. Also, when using multiple smaller units, excess capacity for a given installation can be reduced.

As noted earlier, a low mean time to repair (MTTR) can help improve availability. A system that can be repaired quickly through responsive and certified technicians, readily available parts and easily serviced designs will also improve uptime. Planning for reduced MTTR is a critical component for high availability.

In conclusion, chillers are vital for process cooling, and every application has a different pain threshold for downtime.

A number of approaches are possible for improving the availability of chillers. Depending on the specific situation, the choice can be based upon the economic risks involved. While selection of products with reliable components and good design are at the core of availability planning, redundancy and modularity options can help optimize availability and overall system cost. In addition, having a partner strategy for service, should the need arise, also helps ensure maximum uptime for cooling critical processes.