ESP and boiler commissioning: from air-load curves to the guarantee test

A precipitator that starts badly rarely recovers by tuning. If plates are 8 mm out of alignment in a 300 mm duct, the first field will spark early for the rest of its life; if an insulator tracks on day one, the fault current will be blamed on everything except the insulator; if the guarantee test runs without an agreed O₂ reference, the number it produces is unusable by either party. Commissioning exists to catch each of these at the layer where it is cheap to fix — which is why Arrow Energy Co., Ltd. commissions in a fixed sequence within our boiler and emission-control services scope: cold checks with the casing open, electrical proving on ambient air, hot tuning on gas, and only then the performance test.

The same discipline applies whether we commission our own supply — a new electrostatic precipitator, a rebuild delivered under an ESP upgrade, or a complete train — or act as commissioning engineers for equipment supplied by others. The record produced in the first week becomes the plant's electrical and mechanical baseline for the next twenty years.

Hot commissioning starts only when the boiler delivers stable gas within the precipitator's design window — for bagasse duty typically 150–180 °C at the ESP inlet, never above the 200 °C service limit — because tuning against a swinging inlet condition produces settings that are wrong at every steady state. Under flue gas, the curves change character: space charge from dust raises the voltage the field can hold, and the working point moves. Gas-load V-I curves are recorded per field and compared with air-load; the automatic voltage controllers are then tuned — spark rate setpoints, quench and ramp recovery, intermittent energisation where resistivity demands it — field by field, from inlet to outlet. Opacity is correlated against load and soot-blowing so the plant can later distinguish a real emission change from an operating artefact. Rapping is optimised last, against the opacity trace: intensity and interval per field, slower toward the outlet where re-entrained dust has no downstream field to catch it.

On combined boiler-and-ESP projects, the boiler side is commissioned in parallel and in its own strict order: refractory dry-out on a controlled temperature ramp — indicatively 25–50 °C/h with soak holds per the refractory schedule — so trapped water leaves as vapour rather than as spalled brickwork; safety-valve floating and set-pressure verification; steam-line blowing repeated until target-plate cleanliness acceptance criteria (agreed per project against the turbine or process requirement) are met; and water-chemistry passivation runs to establish the protective magnetite layer before sustained load. A precipitator downstream of a boiler that skipped dry-out or blowing inherits refractory dust, mill scale and temperature excursions in its first week — the week its baseline is being set.

How long does ESP commissioning take?

Indicatively: cold commissioning of a four-field precipitator takes 5–10 days including air-load curves; hot commissioning and controller/rapping optimisation a further 5–10 days as boiler load allows; the performance guarantee test 2–3 days once the unit is stable at the agreed load window. A combined new-boiler-plus-ESP programme runs 4–8 weeks end to end.

Why must the test basis be defined before the guarantee test?

Because dust concentration has no meaning without its basis. 40 mg/Nm³ at 6 % O₂ dry and 40 mg/Nm³ at 11 % O₂ are different emissions by a factor of 1.5. The protocol fixes, in writing and in advance: reference O₂, dry basis, boiler load window, fuel condition, soot-blowing state, port locations and traverse plan. Then both parties read one number the same way.

The test itself is an isokinetic traverse to EPA Method 5 or ISO 9096: sampling velocity matched to duct velocity at each traverse point so the probe neither over- nor under-collects coarse particles, with simultaneous O₂, moisture, temperature and velocity measurement for the correction to reference conditions. For biomass and bagasse duty the reference is typically 6 % O₂ dry, consistent with our published design basis (Electrocyclone + four-field ESP: 24 mg/Nm³ @ 6 % O₂ dry — design basis, illustrative calculation, not a guarantee). Guaranteed figures are stated per project after the technical assessment, on the stated basis.

Deliverables at handover:

• Cold commissioning dossier: alignment records per field, insulator test values, interlock protocols, rapping function reports.
• Air-load and gas-load V-I curve sets per field — the electrical baseline archive.
• AVC and rapping settings register with the tuning rationale per field.
• Opacity–load correlation record.
• Boiler-side records where in scope: dry-out chart, safety-valve floating certificates, steam-blowing target-plate results, passivation chemistry log.
• Performance guarantee test report: raw data, corrections, result on the stated basis.
• Operating and maintenance manual walk-through and operator training record.

The handover file earns its keep for decades. When a field misbehaves in year eight, the maintenance engineer who can lay today's V-I curve over the commissioning curve knows within an hour whether the problem is electrical, mechanical or process-side — and orders the right spare instead of the plausible one. For the same reason we recommend re-recording air-load curves at every major outage: the comparison costs a shift and replaces guesswork with a trend. Plants that lost their original commissioning records can have the baseline reconstructed during a normal shutdown as a stand-alone service.

Commissioning is also where a turnkey emission control project closes out: the same engineers who wrote the design basis run the test that proves it, which removes the most common source of end-of-project dispute — a guarantee written on one basis and tested on another.