Is emi magic?
Electro – Magnetic Interference (EMI) Monitoring may seem like a magic trick. However, the system simply detects conditions associated with EM emissions within the Generator, Isophase Bus, and Transformers. Subject matter experts have been extensively gathering data over the last 40 years. Through their research and testing, they have been able to detect patterns to determine which data sets and waveforms correspond to which failure modes:
- Arcing
- Corona
- Gap Discharge
- Micro – Sparking (rapid gap discharge)
- Partial Discharge
- Random Noise
emi assessment algorithm
Cutsforth™ has created an automated assessment classification algorithm. The tool automatically categorizes time domain waveforms. Each frequency range is associated with certain failure modes. For example, if there is arcing within a 30 MHz – 100 MHz range, plant personnel should look into the Isophase Bus. The system can alert key personnel when EMI increases. Tables are trendable and these features are available in the EMSA platform. This computerized system means that analysis no longer needs to be done manually.
Real – World Example:
A plant in Ohio was monitoring and trending a crack in a winding. They were able to repair the damage during an upcoming planned outage rather than waiting until failure forced a major outage. The riskier approach is to wait until a failure occurs, which can be incredibly expensive. Merely stopping an Isophase Bus failure more than pays for the EMI Monitoring system.
early detection
The first step in analyzing waveforms is to establish a stable base line. Plants can sit at a stable base line for years, before it starts moving and changing, and change is bad. Early detection is greatly important. Trend line changes may be missed if plants are only doing periodic measurements. With continuous monitoring, the computer captures data every 15 minutes trending the information over time. Many failure modes, such as moisture intrusion problems like gasket leaks, are difficult to discover using periodic measurements.
how to compare a power spectrum
The image to the right is a trend platform associated with 8,000 different points that has been demodulated. Of the two sister units, the Blue Line Generator is at the base line, and seeing normal figures. However, the Red Line Generator was left unanalyzed, and had been degrading over years becoming worse and worse. Eventually it had to be repaired forcing an outage.
When the trend value increases, plants need to start looking for certain waveform patterns or data sets to decipher which failures are possible. Since the spectrum degrades slowly over time, the change tends to be subtle. EMI Monitoring helps personnel notice long term trends to discover failures like insulator contamination or Partial Discharge (PD).
financial impacts
Converting one Major Outage to Minor Outage over five years:
Note: the results are illustrative only. Specific risks, costs, and prices will vary. This is not a guarantee of savings nor a guarantee of eliminating risk. Proper maintenance must be performed
conclusion
EMI Monitoring gives plants the capability to measure and trend components from the generator to the GSU over time. Specific failure modes correlate defects into specific frequency ranges. This system provides a long-term planning tool, which optimizes maintenance and capital budgets.
For information on Electro-Magnetic Interference Monitoring click on the image below to watch the Professional Development Webinar.
