A picture is worth a thousand words, especially when inspecting equipment. However, if the proper procedures are not in place to act on what you are seeing, potential problems go unresolved. A good inspection program needs proper tools and documentation. But just as importantly, it also needs a defined process that evaluates risks, benefits and costs so timely action can be taken.

For many years, Tennessee Valley Authority (TVA; Chattanooga, Tennessee, U.S.) has performed substation inspections. The use of infrared began in the late 1980s. By 1991, TVA was using four infrared cameras in substations. In 1991, a bushing failed violently. Old test records revealed an infrared image indicating a hot bushing connection, yet no guidelines were in place to act on the information. From that point forward, infrared became a key predictive-maintenance tool. Soon after, the use of ultrasonic devices became a part of the predictive tool set.

Early in 1993, the first focal plane array infrared cameras were purchased to provide sharper images. Later that year, TVA's Transmission Support and Technology Advancement departments teamed up with the EPRI M&D (Maintenance and Diagnostic) Center. TVA had the technical tools but was missing two essential ingredients — communication and documentation. Therefore, TVA developed a comprehensive program, resulting in the following approach.

Communication is the first step. Each substation survey starts with an entrance meeting. The purpose is to be aware of safety issues, explain the work scope and address any noted substation concerns. After the survey, an exit meeting is held with the local transmission service manager and transmission maintenance coordinator to inform them of any discrepancies. Thorough documentation is the next key ingredient. After a survey, the team generates a report and sends it to the local manager and corporate office. Identifying the potential avoided cost of each discrepancy is a critical part of the report.

In 1998, TVA's Transmission Support department started the Level 2 Inspection (L2i) program as a pilot. It was based on lessons learned from the work with the EPRI M&D Center. TVA formed seven teams, each composed of an electrician from the maintenance side and an engineer from the testing side. TVA surveyed 106 substations in six months with a potential avoided cost of US$4.3 million. Bi-weekly teleconferences were held to discuss better survey techniques and share information. The L2i pilot justified a team in each of TVA's 18 Transmission Service Centers.

Methods and Technologies

TVA looked into using seven different methods/technologies for doing inspections to determine what gave the most investment return. The top four types — visual, infrared, ultrasonic (airborne) and dissolved gas-in-oil analysis (Hydran) — set the new bases for the L2i program.

Investment Return Contribution per Method/Technology

Visual	21% *
Infrared	59% *
Ultrasonic (Airborne) — 40 kHz plus	10% *
Dissolved Gas-in-Oil Analysis	3% *
Ultrasonic (Contact) — 40 kHz to 250 kHz	3%
Sonic — 20 Hz to 20 kHz	2%
Vibration — <1000 kHz	2%

Cost Benefits

To gain program approval, management requires cost benefits. TVA used the following empirical formula to calculate potential avoided costs:

Avoided Cost = (Worst Case Cost)(1%) + (Possible Case Cost)(10%) + (Probable Case Cost)(89%) - Actual Cost to Fix

The weighting factors (percentages) for each case took into account probability and failure severity.

After a period of time, average avoided costs were calculated for each group and used for tracking purposes. Note: Avoided costs are determined only when the problem is fixed; otherwise, identified problems are shown with potential avoided costs.

Communication

Good communication was the glue binding the different teams together and keeping management engaged. The following guidelines were used:

• Conduct an entrance meeting before starting the inspection to review test/maintenance records and discuss areas that might need a closer look.

• Discuss safety concerns.

• Discuss equipment concerns including those out-of-service pieces, which will need inspection at a later date.

• Provide immediate notification of critical discrepancies as soon as they are found.

• Hold an exit meeting after completing the inspection to cover loose ends and set priorities for maintenance activity.

• Hold regular teleconferences with all of the inspection teams to review lessons learned.

• Communicate the cost savings to management.

Documentation

Good documentation follows good communication hand in hand. Seven steps are recommended to provide high-quality reports:

• Develop and document guidelines for cost benefits.

• Use a digital camera and an infrared camera with digital image storage.

• Develop and document guidelines for severity criteria.

• Capture ultrasound data on a digital recorder.

• Use a database for collected information.

• Make hard copies for the maintenance crews to use in the field.

• Document cost savings.

Successful substation inspections require a formal and comprehensive program. The program needs the right equipment and technical expertise to perform the inspections, while standing on good communication and documentation, and headed by cost benefits. Ultimately, the program must provide all of the information necessary to make timely and prudent business decisions.

Mark B. Goff is a staff system engineer for Tennessee Valley Authority (TVA). He received the BSEE degree from the University of Kentucky in 1983 and went to work for TVA as a field test engineer. He joined TVA's transmission staff in 1990 as a lead engineer for substation large power equipment. Since being on the staff, Goff has developed TVA's predictive maintenance program for TVA's substations and is currently working on the development of an online transformer monitoring program for TVA's 500-kV grid. He is register professional engineer in Kentucky.

Two Examples: Discrepancies and Avoided Costs

The oil level can be seen on a transformer bushing. The normal oil level in the bushing is near the top. The oil level in the bushing shown is less than halfway down.

In this example, the qualitative images and actual temperatures are not as significant as the thermal profile.

Low Oil	$1000
Pegged Low (Oil gauge reading below scale)	$2000
Level greater than halfway down in the porcelain	$Cost of the transformer

A breaker does not exhibit iron and copper losses like those seen in transformers. Therefore, a breaker's temperature typically runs slightly greater than ambient. If a temperature rise is seen in a breaker, it must be coming from a high contact resistance inside. Small temperature rises can be significant because the actual contacts are enclosed and the actual contacts cannot be viewed directly. By comparing the tanks, it was noted that the tank on the left was only 10°C (18°F) delta greater than the other two. Closer examination revealed severely burnt contacts that would have led to a breaker malfunction.

In this example, the quantitative images and actual temperatures are as significant as the thermal profile. Temperature rise is not based on ambient. Compare temperatures of all three phases, across the apparatus and with similar devices in like conditions.

Severity Rating	Potential Avoided Cost
Critical >10°C delta	$20,000
Serious 5°C to 10°C delta	$10,000

Definitions of Limits

• Critical

  Immediate investigation with the possibility of removing the equipment from service, depending upon the investigation.

• Serious

  Investigate as soon a possible and increase the frequency of scanning, depending upon the investigation.

• Intermediate

  Trend at regular scanning frequency. In power plant switchyards, multiply the potential avoided costs by factors of four (fossil and hydro) and eight (nuclear), respectively.