That's a weird spec - I've never seen that before. Fan-cooled motors often have a required direction of rotation in order to maintain adequate cooling air flow across the windings.
Yeah, I agree with dpc. There are multiple reasons for choosing directional (axial flow) fans. It may reduce noise. But based on various factors, certain applications will be more efficient as an axial fan.
Realistically if I were tasked to comply with the spec, I'd ignore the phrase "in view of noise abatement method", and for all fans that are unidirectional (for any reason whatsoever) the direction should be marked and should be reversible by flipping the fan.
Axial fans are less noisy and unidirectional and hence motors with such fans should have a DOR marker.
Curved (forward or backward) centrifugal fans are also unidirectional and should have a DOR marker.
Only centrifugal fans with radial blades are bidirectional and most big motors come with them.
Latest German VFD duty motors are very quiet (less than 70 dB in my shop) with externally driven unidirectional axial flow fans thus allowing bidirectional operation of the motor.
It is -occasionally - possible that operating a directional high-speed fan (above 1800 rpm, for example) in the direction opposite to what was intended will give rise to some fairly extreme noise issues due to the difference in the flow created. It is also possible that there is a flat surface with relatively little in the way of bracing that is affected - which may result in a "rumble" kind of noise. For these cases, the "noise reduction" requires that the fan be operated in the correct direction.
My guess is the phrase is a result of less-than-optimal-translation from another language to English.
Converting energy to motion for more than half a century
In the large hydro power generating station where I worked, the unit headgates were hoisted with geared 40 HP 3-phase air-cooled induction motors that turned at just under 1800 rpm; their cooling fans were pitched to provide the necessary air flow during hoisting, which took about ten minutes. The noise level when hoisting was quite tolerable without hearing protection being required.
However: directly coupled to the other end of the hoist motor was a centrifugal fan that spun backwards while hoisting, and that acted as a fly to limit the speed of the descent when a headgate was being lowered; the drop took less than three minutes, and the howl generated was right intolerable. Even with hearing protection on, it was better to just not be there, as the motor line would rotate at speeds in excess of 6,000 rpm.
CR
"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]
Emergency headgate drop in the case of loss of generating unit governor control needs to happen ASAP; the place was built in the 1950's, and to my understanding the need to get the headgates closed quickly trumped any consideration of energy recovery.
Additionally, I'm not an accountant, but the capital expense needed to put this in place for on 48 unit headgates [ three per unit ] to recover three minutes worth of energy for something that happens quite infrequently could in my view never be economically justifiable, probably not even today.
Beyond that, the system as installed has an extremely high level of reliability, seeing as how the both the control circuitry and the application / release mechanism of the headgate brakes is all DC with multiple station batteries available for supply.
CR
"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]
