The generator cooling blower that tripped every time: 0.0 megohms told the whole story

At-a-Glance

The Setup

A generator cooling and ventilation blower is a small fan with a large consequence. On a 2.5 MW turbogenerator set it keeps the generator's air path moving. The duty is modest — 1,100 CMH at 300 mmWC on 3 HP — but availability is not optional: if the GVC blower is down, the packager cannot hand the set over, and on a sugar co-generation plant mid-commissioning a stalled handover stalls the season plan. Commissioning was run by the packager's electrical engineer at site, with our Tech Services engineer supporting from our end.

The Complication

The blower ran cleanly for about a week. Then, one evening, the motor was found tripped. On restart it tripped again — and again. The site's report was blunt: the motor was drawing more than 6.0 A against a rated current of 4.2 A.

When a fan motor pulls high current, the obvious suspect is the fan: an impeller rubbing, a bearing seizing, a damper left wide open so the fan runs out on its curve into the high-flow, high-power corner. "The fan is overloading the motor" is the reflex diagnosis on a new installation — and it has sent many engineers to strip a perfectly good blower.

It was wrong here. Nothing mechanical had changed in that week: same duty, same system resistance, and a full week of healthy running at that operating point. A fan that suddenly needs 45% more current at the same duty is not an aerodynamic story. Something electrical had changed.

The Diagnosis & Fix

We asked for two measurements before anything else. Not a strip-down — two meters.

Ruled out first, on evidence, in this order:

1. Aerodynamic overload. Duty unchanged, damper setting unchanged, no ducting modification reported, and a full week of healthy running at the same operating point. A fan does not silently migrate along its curve while nothing moves.
2. Mechanical drag — bearings, impeller rub, misalignment. No noise, no vibration complaint, no heat, no rub evidence in the site report. Drag that costs 2 A of motor current announces itself audibly long before it trips a starter.
3. The motor windings. This is where the numbers went.

What we found. The site took an insulation-resistance test and a three-phase winding-resistance balance:

• Insulation resistance, Y-phase to ground: 0.0 MΩ. A dead short to earth.
• Winding resistance: R = 10.0 Ω, Y = 8.4 Ω, B = 10.0 Ω. Y low against two healthy phases — the signature of a compromised winding on that phase.

That is not an ambiguous result. A phase shorted to the frame, an asymmetric winding and the consequent over-current draw form a single closed chain. The fan was innocent. The motor — a bought-out, brand-name component still inside its warranty period — had an early-life winding-to-earth failure.

What we did. We characterised the fault fully on site — insulation resistance, phase-resistance balance, running current, and photographs including the motor nameplate. We took ownership of the warranty route rather than leaving the customer to chase it, and instructed a claim against the motor manufacturer quoting the serial number, nameplate data and the measured fault set, so it landed complete rather than bouncing back for information. We did not open the motor ourselves — that is the fastest way to lose a warranty.

The Result

What the record supports:

• We proved where the fault was. 0.0 MΩ Y-phase-to-earth and an 8.4 / 10.0 / 10.0 Ω winding imbalance localise the failure to one phase of one component, with no interpretation required.
• We cleared the blower. Impeller, bearings, casing, alignment and ducting were not implicated by any measurement or observation on record.
• We routed the warranty properly. The claim went to the motor manufacturer with full serial, nameplate and fault-data evidence — the customer did not have to build that file.
• We changed our own process. Pre-dispatch electrical acceptance is now a standing rule on every bought-out motor — phase-to-earth IR, R/Y/B winding-resistance balance, no-load current vs nameplate, recorded on the job traveller — plus a commissioning checklist logging IR and running current at first energisation and after the first run-hours.
• What the record does not confirm. Our file stops where the warranty claim was raised. It does not contain the motor manufacturer's teardown verdict, a replacement-motor test reading, or a re-commissioning sign-off. We will not claim a closure we cannot evidence.

The Takeaway + Call to Action

The transferable lesson: when a fan motor "pulls high current", the first two instruments to reach for are a megger and a milliohm meter — not a spanner. Current is a symptom shared by aerodynamics, mechanics and electrics; only measurement tells you which. Phase-to-earth IR plus a three-phase winding-resistance balance take ten minutes and will either exonerate the motor or convict it, before anyone loosens a bolt. And every bought-out motor that leaves your works untested is a fault you have chosen to let your customer find.

We service centrifugal fans and blowers of any make. Over-current trips, vibration, bearing life, flow shortfall, noise, balancing to ISO 21940, performance verification tested to the IS 4894 / ISO 5801 / AMCA 210 method — if it moves air and it is misbehaving, we will measure it before we quote you a part.

— Jitamitra Electro Engineering · Technical Services

Engineered for Every Application.