A PQM dataset submitted as part of a pre-energisation test pack records supply conditions at the point of measurement, over the monitoring window, under the load conditions present at the time. Whether that constitutes adequate evidence for energisation sign-off depends on how representative those conditions are of the facility's intended operating state.
In a data centre pre-energisation context, the gap between those two things is often significant — and worth examining before the test evidence is accepted as complete.
What the monitoring window captures
A PQM instrument placed at the distribution board or point of common coupling records supply parameters continuously: RMS voltage per phase, phase-to-neutral and phase-to-phase, total harmonic distortion with individual harmonic contributions, neutral current, and frequency. Better instruments capture transient events and voltage dips. Over a sufficiently long monitoring window, this data describes supply behaviour rather than just a point-in-time state.
The question a review needs to consider is whether the window and load conditions in the submitted dataset are representative. Commissioning PQM is typically performed during daytime hours under partial load, with IT infrastructure not yet energised. The supply under those conditions is likely to show lower THD and lower neutral current than it will under the harmonic loading that switched-mode power supplies introduce at full facility operation. A clean PQM dataset captured before the main load source is connected may not be informative about the supply conditions the facility will actually operate under.
Monitoring window adequacy. A 24 to 48 hour minimum captures supply variation through day and night cycles. Seven days gives meaningful visibility if operational load varies by day of week. Short monitoring windows — particularly those under a few hours — may have captured a quiet period that isn't representative of network behaviour.
Load conditions during monitoring. Review should confirm what proportion of design load was present during the monitoring period. Low THD and balanced neutral current at 10% load carries less weight than the same result at or near design load.
Peak values, not averages. Sustained elevated harmonics and peak THD are the relevant indicators, not the average across the window. EN 50160 specifies percentile limits for a reason — brief compliance during a quiet period is not the same as compliance across the operating load range.
Phase asymmetry under unbalanced loading
Phase voltage asymmetry is worth particular attention in facilities with significant single-phase loading distributed across the three phases. Under the load conditions present during commissioning PQM, phase balance may appear well within tolerance. As single-phase IT loads are distributed — potentially unevenly — asymmetry can grow in ways the commissioning snapshot doesn't reveal.
On a recent anonymised data centre review, commissioning PQM showed phase asymmetry well within EN 50160 limits. The monitoring window had been 6 hours under minimal load. When the dataset was assessed against the design load profile, the margin available at full load was considerably narrower than the commissioning result suggested. The asymmetry figure itself wasn't a finding — the monitoring conditions were.
Generator supply and frequency behaviour
Facilities with generator supply introduce a different category of consideration. Grid-connected supply typically holds frequency within tight limits. Generator sets under varying load exhibit frequency excursions as the governor responds to step changes. Sensitive equipment has tolerance limits for frequency variation that may be narrower than the generator's dynamic response under commissioning load transients.
PQM performed only during grid commissioning won't capture generator behaviour. Where the design includes generator transfer as part of normal operation, the test pack should include PQM from generator supply under representative load conditions. Absence of that data is worth noting in a review.
What adequate PQM evidence looks like
Pre-energisation review of PQM evidence is an assessment of whether the dataset supports the conclusion being drawn — that the supply is within tolerance for the equipment to be connected. That assessment requires understanding the monitoring window, the load conditions present, and whether the measurements were taken at a point representative of the supply the facility will operate on. A clean dataset from a quiet commissioning morning is not the same as supply quality confirmation for a fully-loaded critical facility. The review should make that distinction visible before sign-off.