DGA Fixes

by Brett Sargent, LumaSense Technologies.

Solutions are seldom as simple as they should be. The nation’s transmission and distribution infrastructure offers as much proof as any industry.

The average U.S. transformer is more than 40 years old and cannot be replaced fast enough. The same situation exists in many developed countries. How can transformers safely operate longer than their design lives? Also, some 30 percent of U.S. transformer failures are related to load tap changers (LTCs). An obvious solution, one would think, would be to outfit those transformers and LTCs with online monitoring systems designed and proven to diagnose transformer and LTC faults. Simple, right?

Yet, how many U.S. transformers and LTCs have condition-based, online dissolved-gas analysis (DGA)-monitoring systems? Sources indicate less than 5 percent. Even the utilities that own those less than 5 percent offer mixed feedback on the technology’s practicality and effectiveness.

No solution is perfect, but if ever there was an imperfect solution, online DGA embodies it.

After meeting with more than 100 global transformer manufacturers and thousands of utilities during the past decade, we can categorize the DGA issues companies have experienced. But with global assets’ aging and being pushed harder than ever, plus assets’ working beyond their expected lives and fewer qualified workers’ managing these assets, it’s past time to address these issues.

Based on 10 years of feedback, eight main fixes must happen to increase DGA technology adoption and improve its effectiveness in preventing U.S. power outages.

A More Affordable Price

DGA monitors range from $10,000 to $60,000–a small price compared with new, large power transformers that can run several million dollars. If a utility installs equipment on a new, large power transformer, the costs can be capitalized, depreciated and rolled into the new transformer costs, which represents the simplest scenario. In reality, many utilities want to install these devices on transformers in the field that might be more than 30 years old, and the budget is limited. Purchasing a $50,000 instrument for one transformer is difficult to justify. Spending that much for an online monitor for an LTC is even harder, preventing wide-scale implementation across a transformer fleet.

Another problem: The lower the price, the lesser the functionality. Some utilities are not satisfied with buying a $10,000 instrument because their monitoring philosophies are more aligned with a more feature-rich instrument.

“If the price was just half that, I could install quality instrumentation on two transformers instead of one,” they say.

There is also pent-up frustration–DGA instruments have been available many years, but prices have not fallen as expected.

Easier Installation

Unit installation always has been a heavy burden. Installation can range from a half day to five working days and might require building separate stands for the instruments, running additional oil lines to and from the transformer or LTC, complex wiring and even more supports for instruments if they’re attached to a transformer. Sometimes this requires detailed planning and organizing with electrical and mechanical craft labor. Fewer utilities allow installations while transformers are running. Scheduling an outage can be difficult–impossible during summer–especially while pulling in other maintenance during the outage.

The harder the installation, the higher the reluctance. Even when an instrument manufacturer offers to install a unit, there is still reluctance with the scheduling and coordination.

Streamlined Maintenance

The last thing a utility wants is to install instrumentation that requires extensive maintenance. The ideal situation is to use instrumentation that requires no maintenance or calibration, but that’s not always possible. A common acceptable threshold among customers is five years of operation before major maintenance on an instrument. Many customers fear that moving parts such as pumps and filter wheels will need replacements after a few years of service.


A major push exists for industry standardization. As more instrumentation and sensors are used, communicating in a standard format (DNP3.0, IEC61850, Modbus, etc.) is becoming a requirement. Most DGA instrument providers are there, and those that aren’t are moving quickly in that direction. This is becoming increasingly important to overall smart grid strategies.

Next-level Analytics

Any utility will attest that data overload is the biggest downside of using increased instrumentation and sensors on transformers. With an aging work force, utilities want guidance on basics such as understanding readings. They seek a comparison to standards or the use of well-known analytics such as Duval’s triangle, Rogers ratios and key gases. Utilities look to instrumentation to do this for them, vs. interpreting results themselves. Older DGA instrumentation provided only results and numbers with no guidance except an alarm when a certain customer-determined set point was exceeded. As utilities become more strained with resources and overloaded with data, they require DGA instrumentation to provide some type of guidance and next-level analytics for their decision-making. Utilities need DGA instrumentation that will tell them when action is needed, not just inundate them with data.

Operating Environment

Transformers operate in some unfriendly conditions, so DGA instrumentation must cover a wide operating range. Instrumentation specification sheets say instruments can handle the challenges. Nevertheless, utilities have cited that their instruments have frozen, dried and shorted out. Some instrument providers provide special insulation and heating packaging for extreme cold weather and cooling for extreme hot weather, but these come with more cost and maintenance. Utilities cannot take a chance on marginal instrumentation.

Reduced Oil Leakage

Transformers and LTCs typically receive infrequent visits. It might be months between equipment checks, and an oil leak could devastate a transformer or LTC. A DGA instrument must be connected to a transformer or LTC. The more connections, the longer the plumbing runs to and from an instrument, and the type of connections used can all lead to oil leakage. When instrumentation decisions are being made, decision-makers scrutinize how the instruments connect.

Believable Results

A degree of trust surrounds DGA instrument readings. Unfortunately instrument readings don’t always match results from a sample or lab analysis. So which one is right?

The standard procedure for any DGA instrument alarm is for a backup sample to be drawn for verification of the results. The lab wins on the results battle, so there must be a utility trust and confidence level that the DGA instrument provides accurate results, or at least performs consistent measurements. This is a big decision point for utilities–they must be comfortable with what an online instrument tells them. There is a wide array of technology used for these instruments, and established technologies tend to bring utilities more comfort. This does not necessarily mean that the technologies used on transformers and LTCs must have been established for years, but that the technology is well-known and established in the industry. New technologies have a disadvantage in winning utility confidence.

Redefining DGA must start with these eight shortcomings. Doing so provides for increased adoption and higher-performing assets and will modernize the nation’s infrastructure.

We need this solution to be simpler, cheaper and more reliable.

More PowerGrid International Issue Articles
View Power Generation Articles on PennEnergy.com
Previous articleMobile Communication 21st-century Solutions to Meet Market Demands, Mandates
Next articleExecutive Interview: Viewpoints From the Boardroom of Brazil’s Eletrobras

No posts to display