Abnormal noise on a 1200 CMH combustion-air blower: the bearings were innocent

At-a-Glance

The Setup

The customer is an OEM in the boilers & combustion space, also building furnaces, ovens and heat-treatment lines. Our blower leaves their works already integrated into their equipment — so it does not serve their plant, it serves their customers' plants, under their name.

The duty is modest on paper: 1200 CMH at 425 mmWC, 5 HP, 30 °C — a compact, high-static combustion-air blower, small volume and hard pressure. What matters is where it sits: the burner's air supply depends on it, and because it is an OEM-supplied component, a noise complaint on one machine is a question mark over every machine the OEM ships with that blower on it.

The stakes were not "a noisy fan". They were an OEM's confidence in a component carrying their brand.

The Complication

The complaint was one line: noise problem. With it came five phone videos of the blower running.

"Noise" is the least diagnostic word in rotating equipment. It can be aerodynamic (inlet system effect, a badly-fitted flexible connector, duct resonance), mechanical (bearings, coupling, loose fasteners, a fouled impeller), electromagnetic (the motor), or structural (a base that amplifies everything).

Our own first triage went straight to the obvious suspect, and logged itself as a question rather than a finding: "I think this seems bearing noise…?"

Bearings are the obvious suspect because bearings are usually right — and a phone video will happily support that theory, because a video captures only the symptom. Abnormal noise sounds like abnormal noise no matter where in the machine it is born.

This time the obvious suspect was wrong.

The Diagnosis & Fix

We ruled out in the order that costs the customer least.

1. Aerodynamic and system causes. The unit was in normal service at its design duty. No duct modification, damper change or process change coincided with the onset. Nothing pointed to a system-effect or resonance origin.

2. Bearing and fan-side mechanical causes — the leading hypothesis. Here we made a deliberate call: a noise you cannot localise in the field does not get diagnosed in the field. Distinguishing a rub from a bearing from an electromagnetic hum, standing next to a running burner set, is guesswork with a clipboard. So we withdrew the drive motor and brought it back to our works.

3. Factory strip-down — where the answer was. With the motor opened on the bench, the fault was unambiguous: the rotor was contacting the stator. The rub was inside the motor air-gap. The fan bearings — prime suspect all week — were not the source at all.

That is a bought-out component fault, and we treated it as one: the motor was covered under warranty and went for repair. Meanwhile we did the thing that decides whether a warranty case becomes a relationship problem — we issued a spare motor from our own stock, so the blower could be re-assembled and returned to service instead of waiting on the repair cycle.

The Result

What our record proves, and what it does not:

• Proved: the noise originated inside the drive motor, from rotor-to-stator contact, confirmed by strip-down at our works. Not the bearings, not the impeller, not the ductwork.
• Proved: the fault sat in a bought-out component supplied under warranty — and we accepted it as ours to fix rather than pushing it back onto the customer's operation.
• Delivered: motor withdrawn and sent for repair; an interim spare motor issued from our stock; the blower re-assembled on the spare so the customer was not left with a dead machine for the repair duration.
• Changed: the preventive lever from this case is pre-dispatch motor QC — an air-gap check and a no-load run-and-listen on the drive motor as a component, before it goes onto a fan, plus a tighter warranty-return loop with the motor supplier.
• Not confirmed by our record: we hold no documented post-repair noise measurement, no customer sign-off, and no "no recurrence" statement for this case. The record shows the problem, the diagnosis and the corrective action delivered. It stops there, and so do we.

We would rather publish a case that ends with an honest full stop than one that ends with a compliment we cannot evidence.

The Takeaway + Call to Action

Don't stop at the first plausible label. "Bearing noise" is the default hypothesis on any noisy fan because it is usually right — which is exactly what makes it dangerous. Two rules fall out:

1. If you cannot localise the noise in the field, take the component to a bench. A short strip-down beats a long argument.
2. Cover the customer with an interim spare. A repair timeline that turns into a downtime complaint costs far more than the spare would have.

If you have a fan or blower making a noise nobody can explain — vibration you cannot trend, a bearing that keeps getting blamed, a motor nobody has opened — we will come and look at it properly. We service centrifugal fans and blowers of any make, not only our own. Balancing to ISO 21940; vibration assessed against ISO 14694 (rigid BV-3: 4.5 mm/s accept, 7.1 alarm, 9.0 shutdown; flexible 6.3 / 11.8 / 12.5); performance tested to IS 4894 / ISO 5801 / AMCA 210 method.

— Jitamitra Electro Engineering · Technical Services

Engineered for Every Application.