An active harmonic filter corrects the current its CTs show it. Put them in the wrong place and it works hard on the wrong load.
An AHF measures current through its CTs, works out the harmonic content and injects the opposite harmonic current in real time. That is why it suits an EC fan wall, whose load moves with fan speed all day, better than a passive filter tuned to one frequency and one load point. Everything it does depends on what the CTs see.
Three decisions belong on the single-line diagram:
- Which feeder. The CTs must carry the fan wall's current and only the loads you mean to correct. CTs on an incomer that also feeds lighting and chillers hand the filter everything behind it.
- Which side. Most active filters accept CTs on the source side, seeing the residual after correction, or on the load side, seeing the raw load. The filter setting must match, and a swapped CT or phase shows up as a filter that makes distortion worse.
- Where to correct. A filter at each AHU panel follows that load and keeps harmonic current out of the cables upstream. One filter at the main board shares its capacity across loads, but the cables in between still carry the harmonics.
Our design target is THDi ≤ 5% at the point of common coupling. IEEE 519-2022 sets its limits as TDD, by the site's short-circuit ratio, so the harmonic study comes before the order.
Would you correct at each AHU, or at the main board?




