Transformer monitoring: More than meets the eye

In this article:

• Single-gas versus multigas monitoring systems
• How does (and should) monitoring cost factor in?
• Transformer reliability pros focus on criticality
• Isn't every component of an electrical system critical?
• Factors to consider for a criticality analysis

Not all remote transformer monitors are created equal. The insight they provide ranges from a single alarm from a spike in hydrogen – a sudden escalation in the quantity of hydrogen almost certainly signifies an incipient fault – to a complete DGA profile of a piece of equipment that can not only identify faults in the transformer but also identify specific subfaults and help transformer owners and transformer management specialists plan remedial next steps before anyone even sets foot in the substation.

Single-gas versus multigas monitoring systems

Hydrogen monitoring is a key part of transformer fleet management because of its value as the “canary in the coal mine”; however, with the addition of other combustible gases in multigas monitors, the data collected becomes much richer. Monitoring of multiple key gases can tell us what the fault might be – a single gas monitor can indicate only that one exists.

With data points from several combustible gases, transformer owners and transformer management companies can apply industry-standard diagnostic tools such as Roger’s Ratio and Duval’s Triangle (or a combined approach) to understand the specifics of a fault as it happens and to precisely diagnose thermal faults, low- and high-energy discharges, and partial discharges.

• Read "Power risk assessments done right"

Multigas monitors are effective at detecting dangerous faults that may be missed with single-gas monitors. Arcing is not only one of the most destructive faults within a transformer, but also it is one of the most hazardous. Acetylene accompanies arcing and overheating. Often, multigas monitors detect precise levels of acetylene, making it is possible to track thermal faults in a way that is not achievable with the trace amounts of hydrogen tracked by single-gas monitors.

How does (and should) monitoring cost factor in?

There are price differences among monitors, of course; as the precision and diagnostic capability of a monitor increases, so, too, does the price tag. A single-gas hydrogen monitor is, as you would expect, a much smaller investment than, say, a nine-gas monitor that provides enough data to tell a vivid story about a transformer’s health (or, as we see more and more, confuse even the most competent engineer if he or she doesn’t have the right tools to manage the data). Cost is often a deciding factor for smaller organizations or for companies with less financially robust maintenance or reliability programs.

Transformer reliability pros focus on criticality

The initial investment is far from the only factor that should be considered when deciding what kind of remote monitor to use. When we deal with customers who are seasoned reliability and maintenance professionals, it is clear that one aspect of reliability colors every purchasing decision: criticality.

Defining the criticality of a particular piece of equipment in any system (and, it seems, especially within an electric power system) presents several challenges. To begin, however, it’s important to understand what is meant by “criticality” in this context and how ignoring this vital consideration can lead to unexpected failure or overspending – two things that run counter to the business goals of every private company.

Isn't every component of an electrical system critical?

Power systems consist of interconnected assets, with the reliability and resilience of each asset affecting the overall reliability and resilience of the entire system. If a breaker or bushing fails, for example, that failure causes problems for the entire system, and that problem that cannot be rectified until a replacement part is sourced, paid for, shipped, received, and installed. It stands to reason, then, that each component of the electric power system could be considered highly critical in some sense. If one asset fails, the equipment down the line either fails to operate effectively or shuts down completely. This spells inefficiency or downtime – both of which are unacceptable in modern manufacturing.

Transformers can be difficult to assign criticality to because standard criticality analysis tools can be difficult to implement. Determining the impact of failure versus the probability of failure is challenging without at least some prior experience in failure and failure modes. If a transformer is half a decade old, for example, then the chances are good that there is not enough failure data on that model to even guess as to whether it will last a few more months or 50 years.

Transformers can be tricky to replace. In generation systems, backup transformers are often available for speedy replacement in an emergency. If a failure occurs, remediation is a matter of removing the faulty equipment and switching it out for a similar piece of equipment that will be subject to similar loads and stresses. In industrial applications, however, there are constraints that make this sort of preparation impossible. The uniformity found in preplanned transmission systems is not found in manufacturing substations, where high-voltage equipment is specific to the load required for a particular application. Many CFOs would balk at stocking a backup transformer because of the sheer cost involved. When there is no economy of scale, a capital purchase of that magnitude makes a dent in the bottom line. And if the same overbuilt transformer has been working well for decades, there is no exigence for emergency backup. It is always expected to work.

Transformer failure, then, is a complex, time-consuming, and expensive task. New equipment is expensive to construct and to ship, and long lead times mean extended periods of downtime for whichever line is powered by that transformer. Although it is correct to say that every component of an electrical system is critical, the recovery time for transformer failures can be more devastating and can potentially be measured in weeks or months and millions of dollars.

• Read "Why motor monitoring matters so much"

So, what makes some transformers more “critical” than others? One of the most common (and disingenuous) explanations is as follows: The transformer that powers the single fluorescent bulb in an abandoned parking lot is less critical than the transformer that powers a neonatal ward. In real life, of course, those dichotomies don’t exist, but it can help to think in those terms. Obviously, the old parking lot would receive the less-expensive option – the single-gas monitor – and the neonatal facility would get the full DGA capabilities of the nine-gas monitor because the cost of failure is so much higher.

Factors to consider for a criticality analysis

A criticality analysis that is performed on a transformer-by-transformer basis takes more than application into account. The analysis should also consider:

• The current state of the transformer. If a transformer is 20 years old, has never shown signs of a fault, and continues to operate effectively, then less vigilance is required, and a single-gas monitor may provide all the insight needed to maintain the transformer’s current condition. If a transformer is already gassing and in poor health, then a more-frequent and in-depth assessment is required to ensure that the problem doesn’t get worse. A hydrogen monitor does not provide the nuance required to trend the worsening condition of a fault. They also present a practical challenge when setting alerts or alarm limits because in a bid to provide early detection of low-temperature faults, parameters may be set too low, resulting in in “false alarms” where natural variation in gas concentration or stray gassing may trigger an alert or alarm without a fault condition.
• The age, make, and rating of the transformer. Many well-maintained transformers live way past their predicted lifetimes, and many others have failed either randomly or through degradation or lack of maintenance. It is important to understand whether the transformer you own has any known issues or susceptibilities. This will help you determine the likelihood of failure. A transformer management company or a consultant specializing in transformers will probably have access to this data and can help you make a decision on which monitor is likely to be the best fit for your needs.

Transformers are deceptively simple pieces of equipment that require sophisticated tools to maintain correctly. Sometimes, the cost of maintaining that transformer outruns the value that the transformer provides. This doesn’t happen in industrial settings often, but that one powering the abandoned parking lot with the fluorescent bulb? Perhaps that requires only annual oil testing and inspection.

Whether you opt for a single- or multigas monitor for your transformer, the choice isn’t among “good,” “better,” and “best.” Rather, it is a matter of which monitor will be the best fit for your equipment.