Start with the question
A campaign without an objective produces gigabytes of data and no answers. First define what you want to know: is a criterion being met at the point of interconnection? What is causing a line to stop? How is consumption distributed? The question determines the measurement point, the parameters and the criterion against which the results will be compared.
Planning steps
- Know the facility. Single-line diagram, connection type (wye with neutral, delta), voltage levels, main loads and their operating schedule.
- Choose the point or points. Service entrance for compliance; sensitive load feeder for effects; suspect load feeder to identify the origin.
- Select parameters and configuration. Voltage, current, power, harmonics, unbalance, flicker, frequency, events. Transformation ratios, connection type and event thresholds are configured based on nominal voltage.
- Define intervals. An aggregation interval for trends and, separately, event and waveform capture.
- Size the sensors. Clamp or coil range to match the actual current, to avoid measuring at the low end of their scale.
- Set the duration. It must span complete production cycles, including shifts, weekends and operation with and without certain loads, since the phenomena may be intermittent.
Before leaving the instrument
Check that the connections are correct: phase sequence, current sensor polarity, matching between voltage and current phases, and consistency of values (powers with the expected sign, a reasonable phasor diagram). A connection error invalidates the entire campaign. Also synchronize the instrument’s clock with the plant’s records to correlate events.
At the end
Check that there are no gaps in the data, separate atypical periods and evaluate against the criterion agreed in advance, using the statistics it requires. Connections in energized panels must be made by qualified personnel, following the applicable electrical safety regulations.