1、 Reference number ECMA TR/12:2009 Ecma International 2009 ECMA TR/99 2nd Edition / December 2010 Constant Sound Power Fan Curves for Small Air-moving Devices COPYRIGHT PROTECTED DOCUMENT Ecma International 2010 Ecma International 2010 i Contents Page 1 Scope 1 2 References . 1 3 Terms and definition
2、s . 1 4 Abbreviations . 1 5 Iso-acoustic fan curves . 2 5.1 Overview . 2 5.2 Airflow chamber testing 2 5.3 Acoustic plenum testing . 2 5.4 Flow rate determination 3 5.5 Iso-acoustic fan curve determination 7 5.6 Comparison to manual iso-acoustic data . 9 5.7 Iso-acoustic fan curve test report 10 6 I
3、nlet restriction fixture . 11 6.1 Overview . 11 6.2 Notebook inlet restriction fixture . 11 6.3 Comparison to in-system noise . 13 6.4 Iso-acoustic fan curve test report with inlet restriction fixture 14 Annex A (informative) Sample Data 15 ii Ecma International 2010 Ecma International 2010 iii Intr
4、oduction Fan performance in terms of flow and pressure can be measured using an airflow chamber in accordance with AMCA 210, and such a fan curve is normally provided in a manufacturers datasheet. The datasheet may also contain a measured sound pressure level, typically at 1 m from the fan inlet, wi
5、th the fan in free space. Sound power level in a loaded condition, as a function of static pressure, may be measured in accordance with ISO 10302-1 by using an acoustic fan plenum. Such data is much more indicative of the noise the fan will make when installed in a device, but is difficult to compar
6、e to the fan curve without cross-plotting between multiple data sets. A simpler way to comprehensively state fan performance is with a constant sound power, or iso-acoustic, fan curve. Using this iso-acoustic fan curve, fan performance may be easily compared in terms of flow, pressure, and noise. Fo
7、r example, iso-acoustic fan curves for two fans at the same sound power level may be compared to the impedance curve of a system to determine which fan will provide more airflow in that system for a fixed acoustic limit. Alternately, if a sound power level is chosen for the iso-acoustic fan curve th
8、at is acceptable from an ergonomics perspective, the system designer can be confident that the acoustic limit will automatically be satisfied no matter what the system impedance turns out to be in the final design. Inlet effects can have a large impact on fan flow rate and noise generation. For exam
9、ple, an axial fan commonly has a finger guard or grille, while a blower used in a notebook computer operates within a very constrained space. Since these effects cannot be accounted for by a simple pressure drop due to the complicated flow physics, it is desirable that the fan curve for a specific a
10、pplication include such inlet losses directly. A fixture to approximate the inlet restriction in a notebook computer is described that has been shown to correlate closely with the noise of actual systems. This Ecma Technical Report has been adopted by the General Assembly of December 2010. iv Ecma I
11、nternational 2010 “DISCLAIMER This document and possible translations of it may be copied and furnished to others, and derivative works that comment on or otherwise explain it or assist in its implementation may be prepared, copied, published, and distributed, in whole or in part, without restrictio
12、n of any kind, provided that the above copyright notice and this section are included on all such copies and derivative works. However, this document itself may not be modified in any way, including by removing the copyright notice or references to Ecma International, except as needed for the purpos
13、e of developing any document or deliverable produced by Ecma International (in which case the rules applied to copyrights must be followed) or as required to translate it into languages other than English. The limited permissions granted above are perpetual and will not be revoked by Ecma Internatio
14、nal or its successors or assigns. This document and the information contained herein is provided on an “AS IS“ basis and ECMA INTERNATIONAL DISCLAIMS ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY OWNERSHIP R
15、IGHTS OR ANY IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.“ Ecma International 2010 1 Constant Sound Power Fan Curves for Small Air-moving Devices 1 Scope This Ecma Technical Report specifies a method to generate iso-acoustic fan curves for small air-moving devices, with
16、 or without the presence of a fixture to provide a specific inlet condition to the unit under test. A fixture to provide inlet restriction applicable to blowers used in cooling of notebook computers is also described. It is assumed that users are familiar with aerodynamic fan performance measurement
17、s and the use of the acoustic fan plenum to measure sound power of a fan under load. 2 References For dated references, only the edition cited applies. For undated references, the latest edition of the referenced document (including any amendments) applies. ISO 10302-1, Acoustics - Measurement of No
18、ise and Vibration of Small Air-Moving Devices, Part 1: Airborne Noise Emission AMCA 210, Laboratory Methods of Testing Fans for Aerodynamic Performance Rating 3 Terms and definitions For the purposes of this document, the following terms and definitions apply. 3.1 iso-acoustic fan curve relation bet
19、ween pressure and flow performance of a fan along an isobel 3.2 isobel contour of constant sound power level 3.3 airflow chamber device used to measure the flow and pressure performance of an air-moving device 4 Abbreviations AFC Airflow Chamber RPM Revolutions Per Minute 2 Ecma International 2010 5
20、 Iso-acoustic fan curves 5.1 Overview The iso-acoustic fan curve procedure applies to any air-moving device measured on the acoustic plenum. The term “fan” will be used generically to describe the air-moving device. Two measurements must be combined to generate the iso-acoustic curve, one data set f
21、rom an airflow chamber (AFC) and one data set from an acoustic plenum. The AFC data comprises at least one constant voltage fan curve. The acoustic plenum data comprises a set of measurements spanning a wide range of operating points. The AFC data is used to determine the flow rate for each acoustic
22、 plenum measurement by an interpolation technique. The flow, pressure, and noise data is then scaled, using a chosen exponent, to a desired sound power target and a final fitting function generated to describe the iso-acoustic curve. NOTE If an inlet restriction fixture is used (see Clause 6), it sh
23、ould be used for both the AFC and acoustic plenum testing. 5.2 Airflow chamber testing The minimum requirement is to measure at least one constant voltage fan curve, recording at least flow, pressure, and fan speed data. The recommended procedure is to measure at two voltages (60% and 100% of the ra
24、ted voltage) with at least eight points on each fan curve. An example data set is shown in Figure 1. Once installed on the AFC, it is estimated that each data point should take one minute to collect. 0204060801001201401601802000 1 2 3 4 5 6 7 8Flo w (m3/h)Pressure (Pa)5 V3 VFigure 1 Constant voltage
25、 fan curves (QV, PV) 5.3 Acoustic plenum testing The acoustic plenum data should span a wide range of operating points, from maximum flow to near stagnation pressure, at a minimum of two voltages. The exact voltages and pressure values are not important; no effort should be spent to match any operat
26、ing point measured in 5.2. The minimum requirement is to measure five operating points at each of two voltages, recording A-weighted sound power level, plenum Ecma International 2010 3 pressure, and fan speed. The recommended procedure is to measure at least six operating points for each of three vo
27、ltages (60%, 80%, and 100% of the rated voltage). An example data set is shown in Figure 2. Once installed on the acoustic plenum, it is estimated that each data point should take one to two minutes to collect (longer if the slider on the plenum must be manually operated). NOTE Due to plenum leakage
28、, the maximum pressure on the acoustic plenum may be less than the stagnation pressure recorded on the AFC at the same voltage. If the maximum flow of the fan is less than twice the leakage rate of the plenum, the fan should be considered too small to measure using the plenum. 4445464748495051525354
29、550 20 40 60 80 100 120 140 160Pr essure (Pa)Sound Power (dBA)5 V4 V3 VFigure 2 Acoustic plenum data (LWP, PP) NOTE Example point for which the flow rate is to be determined is circled. The fan speed is 6 304 RPM. 5.4 Flow rate determination For each operating point measured in 5.3, such as the one
30、circled in Figure 2, the flow rate on the acoustic plenum must be determined. Although this can be achieved by manually matching the fan speed and plenum pressure on the AFC, the recommended procedure is to interpolate based on the constant voltage data from 5.2. For each acoustic plenum operating p
31、oint, the fan speed NP and plenum pressure PP are known, while the flow rate QP is desired. First, all of the constant voltage fan speed NV, AFC pressure PV, and flow rate QV data points are scaled to NP using the following equations, where subscript P indicates measured plenum data, subscript V ind
32、icates the constant voltage fan curve data, and subscript S indicates the data is scaled to a constant speed: VPVS NNQQ (1a) 4 Ecma International 2010 2VPVS NNPP (1b) An example data set is shown in Figure 3, where each point is a (QS, PS) pair for the same NP, 6 304 RPM in this case, based on the d
33、ata in Figure 1 and the fan speed for the indicated point in Figure 2. Second, the data in Figure 3 is then modified by swapping the horizontal and vertical axes, as shown in Figure 4, and a smoothing function applied. 0204060801001200 1 2 3 4 5 6 7 8Flo w (m3/h)Pressure (Pa)Scal ed t o 6 304 RPMFig
34、ure 3 Constant voltage data scaled to acoustic plenum fan speed (QS, PS) 0123456780 20 40 60 80 100 120Pr essure (Pa)Flow(m3/h)Scal ed t o 6 304 RPMFigure 4 Constant fan speed data with swapped axes (PS, QS) Ecma International 2010 5 In order to make a smooth function out of the experimental data in
35、 Figure 4, a general polynomial least-squares regression is recommended. A spline is used here for demonstration. Engineering judgment must be used to select the stiffness of the spline. A good stiffness is shown in Figure 5, while an overly stiff spline is shown in Figure 6, and an overly flexible
36、spline is shown in Figure 7. 0123456780 20 40 60 80 100 120Pr essure (Pa)Flow(m3/h)Scal ed t o 6 304 RPMSpline FitFigure 5 Spline fit to constant fan speed data with swapped axes (PS, QS) 0123456780 20 40 60 80 100 120Pr essure (Pa)Flow(m3/h)Scal ed t o 6 304 RPMSpline FitFigure 6 Overly stiff splin
37、e 6 Ecma International 2010 0123456780 20 40 60 80 100 120Pr essure (Pa)Flow(m3/h)Scal ed t o 6 304 RPMSpline FitFigure 7 Overly flexible spline Third, once the spline fit is established, the desired flow rate QP is simply obtained by entering the known acoustic plenum pressure PP and reading the co
38、rresponding flow rate from the spline. For example, as shown in Figure 8, for an NP of 6 304 RPM and PP of 78,6 Pa, corresponding to the circled point in Figure 2, the flow rate on the acoustic plenum QP is estimated to be 3,54 m3/h. This procedure is repeated for each operating point on the acousti
39、c plenum in Figure 2. Once implemented in software, the spline fit procedure to determine flow rate is extremely fast to execute; flow rates can be determined as quickly as fan speeds and pressure values can be entered. 0123456780 20 40 60 80 100 120Pr essure (Pa)Flow(m3/h)Scal ed t o 6 304 RPMSplin
40、e FitFigure 8 Estimated flow rate for acoustic plenum at known fan speed and pressure Ecma International 2010 7 5.5 Iso-acoustic fan curve determination After the procedure in 5.4 has been completed, the flow rate, plenum pressure, fan speed, and sound power values are all known for each operating p
41、oint measured on the acoustic plenum according to 5.3. In order to scale this data to any desired constant sound power level, a new fan speed for each point is first determined, using the following equation, where a subscript P indicates the original acoustic plenum data point and a subscript T indi
42、cates the data point scaled to the desired sound power level target LWT, in A-weighted decibels: E LL WWNN PT 1,01,0PT 10 (2) Here E is the scaling exponent (described below). Once NT is known, the flow and pressure are scaled to this fan speed: PTPT NNQQ (3a) 2PTPT NNPP (3b) For example, again usin
43、g the indicated data point in Figure 2 and the results of 5.4, NP is 6 304 RPM, PP is 78,6 Pa, QP is 3,54 m3/h, and LWP is 49,7 dBA. Applying (2) and (3) with an LWT of 50 dBA and E of 6,0 results in NT of 6 377 RPM, QT of 3,58 m3/h, and PT of 80,5 Pa. The set of all points (QT, PT) then represents
44、an estimate of the iso-acoustic fan curve for a given LWT. A fitting function, typically a low-order polynomial, is then used to describe the final iso-acoustic curve, as shown in Figure 9 for a sound power level of 50 dBA and scaling exponent of 6,0. The scaling exponent is chosen to minimize the s
45、pread in the (QT, PT) scaled data, and this is why acoustic plenum data must be collected at a minimum of two different voltages. The scaling exponent is normally in the range of 5,0 to 7,0 depending on the fan design. A value between 5,0 to 6,5 is common for blowers used in cooling of notebook comp
46、uters. y = -4, 4549x2+ 4, 7757x + 119,15R2= 0, 99540204060801001201401600 1 2 3 4 5 6 7Flo w (m3/h)Pressure (Pa)Measure d d ataScal ed t o iso -aco ust icFigure 9 Scaled 50 dBA iso-acoustic fan curve (E = 6,0) 8 Ecma International 2010 In order to demonstrate the effect of the scaling exponent and v
47、arious sound power targets, consider the following figures. Figure 10 shows a scaling exponent of 5,0, while Figure 11 shows a scaling exponent of 7,0, both at a sound power level target of 50 dBA. It can be seen that the scaling exponent exerts some influence on the iso-acoustic curve, but the fina
48、l result is not dramatically different due to the averaging effect of the fitting function. Figure 12 shows a sound power level target of 45 dBA, while Figure 13 shows a sound power level target of 55 dBA, both with a scaling exponent of 6,0. At 45 dBA, the resulting iso-acoustic curve is on the low
49、er edge of the measured data, while at 55 dBA the curve is beyond the measured data. Since acoustic plenum data was taken at 100% of the rated voltage, this fan would not be expected to reach the 55 dBA iso-acoustic fan curve under normal circumstances. Caution should also be used when estimating iso-acoustic curves below the measured data as the fan may not operate in a stable fashion at low speeds. y = -4, 184x2+ 2, 8945x + 123,29R2= 0, 97550204060801001201401600 1 2 3 4 5 6 7Flo
