Please use this identifier to cite or link to this item:
|Title:||Aerodynamic and vibroacoustic noise of small axial-flow fans : analysis and control|
|Subjects:||Hong Kong Polytechnic University -- Dissertations|
Fans (Machinery) -- Noise
|Publisher:||The Hong Kong Polytechnic University|
|Abstract:||Ventilation fans are widely used in all kinds of consumer products. Hong Kong has a high capacity of manufacturing such consumer products with many factories in the Pear River Delta region. As the living standard improves, the problem of fan noise emission from these products becomes more urgent than the performance of the ventilation fans. Now there is a need to improve the understanding of fan noise and to offer strategies for reducing such noise. This is the motivation behind the current project. Fan noise has been studied for over half a century and many mechanisms have been understood. In some cases, they are also satisfactorily quantified. But, to the best of such knowledge, much remains to be investigated, both qualitatively and quantitatively. The problem of noise radiation from large turbomachines in aerospace applications has received a lot of attention for obvious reasons. The same is not true for small ventilation fans. Noise from small fans shares many features with that of large turbomachines, but there are important differences. For instance, a turbo-fan aircraft engine may have a diameter of nearly 1m and the gap between the rotating blades and the stationary casing, i.e. the tip clearance, is about 1mm. The flow in this gap is believed to cause a lot of energy loss as well as noise. A much smaller consumer product fan normally has a tip clearance of 2mm. The problem of noise radiation from this region becomes much more severe. The big clearance is required for safety reasons. The issue of safety comes up for consumer products instead of the aircraft because of the low accuracy, hence low cost, at which such fans are manufactured. Because of the need to keep the cost low, the motor assembly often vibrates fiercely during rotation. This causes a lot of extra noise. This forms a special class of technical problem which is the focus of this project. Specifically, the project aims to separate the component of noise radiated from the body vibration of the fans, called vibrational noise, from the aerodynamic noise caused by the unsteady flow features. Based on the finding of this noise identification, a technical recommendation will be made to reduce the overall noise. A second objective of the project is to test a few preliminary noise abatement measures for the aerodynamic noise, having identified and excluded the vibrational noise from analysis.|
The difference in the directivity of the two types of noise is utilized to achieve the purpose of noise separation. The measurement focuses on the detailed history of sound pressure, p(t). The radiation of aerodynamic noise is assumed to be symmetrical with respect to the rotational axis. For an observer standing in the wake of the air flow, the noise measured at a left-hand side point, pa₁(t), should be identical to the noise measured at a mirror point on the right-hand side, Pa₂(t)=Pa₁(t). On the other hand, noise radiated by the vibration of the motor assembly and the blades will feature the opposite phase relationship of pv₁(t)=-pv₂(t). Here subscripts 'a' and 'v' stand for 'aerodynamic' and 'vibrational', respectively. The vibration is assumed to be limited on the rotational plane. The components of aerodynamic noise, pa, and the vibrational noise, Pv, can thus be identified by the measurements of the total noise at the two sides. If the readings at the two symmetrical points are P₁ and P₂, then the aerodynamic noise and vibration noise can be separated by pa=(P₁+p₂)/2, PV=(P₁-P₂). The error of such noise separation will be attributed to the assumptions on the nature of aerodynamic and vibrational noises. In order to assess the accuracy of these assumptions, an accelerometer is attached to the motor assembly and the vibration signal is used for a correlation study with respect to Pa and Pv. It is found that the correlation with Pv is indeed much higher than that with Pa. During the analysis, it is also found that the friction caused by the five brushes inside the motor is particularly noisy, and the filtering of this signal from P₁ and P₂ improves the contrast of the noise correlation. Attempts are also made to reduce the aerodynamic noise and the total noise. First, the tip clearance is reduced from 2-3mm to 1-2mm by using a flexible membrane which can be regarded as safe for the consumer products. The result shows significant improvement for the air flow capacity under the same rotational speed. The rotational speed can thus be reduced to achieve the same ventilation performance, and the noise is reduced by several decibels. Second, a layer of densely perforated casing is added to absorb the fan noise. The result shows more than 5dB reduction at the peak noise frequency and moderate reduction is also achieved for the overall noise.
|Description:||102 leaves : ill. ; 30 cm.|
PolyU Library Call No.: [THS] LG51 .H577M ME 2002 Chong
|Rights:||All rights reserved.|
|Appears in Collections:||ME Theses|
PolyU Electronic Theses
Files in This Item:
|b16378489_link.htm||For PolyU Users||162 B||HTML||View/Open|
|b16378489_ir.pdf||For All Users (Non-printable)||3.42 MB||Adobe PDF||View/Open|
All items in the PolyU Institutional Repository are protected by copyright, with all rights reserved, unless otherwise indicated. No item in the PolyU IR may be reproduced for commercial or resale purposes.